Uploaded by Jia Xin

Steel Construction and Design Manual - 6th ed BS

Full Page Ad, 3mm Bleed
Steel Designers' Manual - 6th Edition (2003)
Contents
Introduction to the sixth edition
Contributors
Notation
xi
xv
xxv
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
SECTION 1: DESIGN SYNTHESIS
1
Single-storey buildings
Range of building types; Anatomy of structure; Loading; Design of
common structural forms
1
2
Multi-storey buildings
Introduction; Factors influencing choice of form; Anatomy of structure;
Worked example
42
3
Industrial steelwork
Range of structures and scale of construction; Anatomy of structure;
Loading; Structure in its wider context
94
4
Bridges
Introduction; Selection of span; Selection of type; Codes of practice;
Traffic loading; Other actions; Steel grades; Overall stability and
articulation; Initial design; Worked example
124
5
Other structural applications of steel
Towers and masts; Space frames; Cable structures; Steel in residential
construction; Atria
169
SECTION 2: STEEL TECHNOLOGY
6
Applied metallurgy of steel
Introduction; Chemical composition; Heat treatment; Manufacture
and effect on properties; Engineering properties and mechanical tests;
Fabrication effects and service performance; Summary
222
7
Fracture and fatigue
Fracture; Linear elastic fracture mechanics; Elastic–plastic fracture
mechanics; Materials testing for fracture properties; Fracture-safe
design; Fatigue
248
8
Sustainability and steel construction
275
Introduction; Economic impacts; Social impacts; Environmental impacts;
Embodied energy; Operational energy; Summary
iii
Steel Designers' Manual - 6th Edition (2003)
iv
Contents
SECTION 3: DESIGN THEORY
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
9
Introduction to manual and computer analysis
Introduction; Element analysis; Line elements; Plates; Analysis of
skeletal structures; Finite element method
286
10
Beam analysis
Simply-supported beams; Propped cantilevers; Fixed, built-in or
encastré beams; Continuous beams; Plastic failure of single members;
Plastic failure of propped cantilevers
325
11
Plane frame analysis
Formulae for rigid frames; Portal frame analysis
342
12
Applicable dynamics
Introduction; Fundamentals of dynamic behaviour; Distributed
parameter systems; Damping; Finite element analysis; Dynamic testing
354
SECTION 4: ELEMENT DESIGN
13
Local buckling and cross-section classification
Introduction; Cross-sectional dimensions and moment–rotation
behaviour; Effect of moment–rotation behaviour on approach to design
and analysis; Classification table; Economic factors
373
14
Tension members
Introduction; Types of tension member; Design for axial tension;
Combined bending and tension; Eccentricity of end connections; Other
considerations; Cables; Worked examples
383
15
Columns and struts
Introduction; Common types of member; Design considerations;
Cross-sectional considerations; Compressive resistance; Torsional and
flexural-torsional buckling; Effective lengths; Special types of strut;
Economic points; Worked examples
402
16
Beams
Common types of beam; Cross-section classification and moment
capacity, Mc; Basic design; Lateral bracing; Bracing action in bridges –
U-frame design; Design for restricted depth; Cold-formed sections as
beams; Beams with web openings; Worked examples
431
17
Plate girders
Introduction; Advantages and disadvantages; Initial choice of crosssection for plate girders in buildings; Design of plate girders used in
buildings to BS 5950: Part 1: 2000; Initial choice of cross-section for
plate girders used in bridges; Design of steel bridges to BS 5400: Part 3;
Worked examples
470
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Contents
v
18
Members with compression and moments
Occurrence of combined loading; Types of response – interaction;
Effect of moment gradient loading; Selection of type of cross-section;
Basic design procedure; Cross-section classification under compression
and bending; Special design methods for members in portal frames;
Worked examples
511
19
Trusses
Common types of trusses; Guidance on overall concept; Effects of
load reversal: Selection of elements and connections; Guidance on
methods of analysis; Detailed design considerations for elements;
Factors dictating the economy of trusses; Other applications of trusses;
Rigid-jointed Vierendeel girders; Worked examples
541
20
Composite deck slabs
Introduction; Deck types; Normal and lightweight concretes; Selection
of floor system; Basic design; Fire resistance; Diaphragm action; Other
constructional features; Worked example
577
21
Composite beams
Application of composite beams; Economy; Guidance on span-todepth ratios; Types of shear connection; Span conditions; Analysis of
composite section; Basic design; Worked examples
601
22
Composite columns
651
Introduction; Design of encased composite columns; Design of concretefilled tubes; Worked example
SECTION 5: CONNECTION DESIGN
23
Bolts
Types of bolt; Methods of tightening and their application; Geometric
considerations; Methods of analysis of bolt groups; Design strengths;
Tables of strengths
671
24
Welds and design for welding
685
Advantages of welding; Ensuring weld quality and properties by the
use of standards; Recommendations for cost reduction; Welding
processes; Geometric considerations; Methods of analysis of weld groups;
Design strengths
25
Plate and stiffener elements in connections
Dispersion of load through plates and flanges; Stiffeners; Prying forces;
Plates loaded in-plane
711
26
Design of connections
Introduction; Simple connections; Moment connections; Summary;
Worked examples
721
Steel Designers' Manual - 6th Edition (2003)
vi
27
Contents
Foundations and holding-down systems
Foundations; Connection of the steelwork; Analysis; Holding-down
systems; Worked examples
816
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
SECTION 6: OTHER ELEMENTS
28
Bearings and joints
Introduction; Bearings; Joints; Bearings and joints – other
considerations
842
29
Steel piles
Bearing piles; Sheet piles; Pile driving and installation; Durability
867
30
Floors and orthotropic decks
Steel plate floors; Open-grid flooring; Orthotropic decks
906
SECTION 7: CONSTRUCTION
31
Tolerances
Introduction; Standards; Implications of tolerances; Fabrication
tolerances; Erection tolerances
917
32
Fabrication
Introduction; Economy of fabrication; Welding; Bolting; Cutting;
Handling and routeing of steel; Quality management
948
33
Erection
Introduction; The method statement; Planning; Site practices; Site
fabrication and modifications; Steel decking and shear connectors;
Quality control; Cranes and craneage; Safety; Special structures
971
34
Fire protection and fire engineering
Introduction; Standards and building regulations; Structural
performance in fire; Developments in fire-safe design; Methods of
protection; Fire testing; Fire engineering
1013
35
Corrosion and corrosion prevention
The corrosion process; Effect of the environment; Design and
corrosion; Surface preparation; Metallic coatings; Paint coatings;
Application of paints; Weather-resistant steels; The protective
treatment specification
1030
36 The Eurocodes
The Eurocodes – background and timescales; Conformity with EN
1990 – basis of design; EC3 Design of steel structures; EC4 Design
of composite steel and concrete structures; Implications of the
Eurocodes for practice in the UK; Conclusions
1053
Steel Designers' Manual - 6th Edition (2003)
Contents
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Appendix
Steel technology
Elastic properties of steel
European standards for structural steels
Design theory
Bending moment, shear and deflection tables for
cantilevers
simply-supported beams
built-in beams
propped cantilevers
Bending moment and reaction tables for continuous beams
Influence lines for continuous beams
Second moments of area of
two flanges
rectangular plates
a pair of unit areas
Geometrical properties of plane sections
Plastic modulus of
two flanges
rectangles
Formulae for rigid frames
Element design
Explanatory notes on section dimensions and properties, bolts and welds
1 General
2 Dimensions of sections
3 Section properties
4 Bolts and welds
Tables of dimensions and gross section properties
Universal beams
Universal columns
Joists
Universal bearing piles
Hot-finished:
circular hollow sections
square hollow sections
rectangular hollow sections
Cold-formed:
circular hollow sections
square hollow sections
rectangular hollow sections
Asymmetric beams
Parallel flange channels
vii
1071
1072
1077
1079
1087
1094
1102
1105
1116
1118
1122
1124
1127
1128
1130
1148
1149
1151
1160
1166
1172
1175
1178
1181
1183
1185
1187
1190
1192
1195
1197
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
viii
Contents
Two parallel flange channels:
laced
back to back
Equal angles
Unequal angles
Equal angles: back to back
Unequal angles: long legs back to back
Castellated universal beams
Structural tees cut from universal beams
Structural tees cut from universal columns
Extracts from BS 5950: Part 1: 2000
Deflection limits (Section two: Table 8)
Design strengths for steel (Section three: Table 9)
Limiting width-to-thickness ratios for sections other than CHS and
RHS (Section three: Table 11)
Limiting width-to-thickness ratios for CHS and RHS (Section three:
Table 12)
Bending strengths (Section four: Tables 16 and 17)
Strut table selection (Section four: Table 23)
Compressive strength (Section four: Table 24)
Connection design
Bolt data
Hole sizes
Bolt strengths
Spacing, end and edge distances
Maximum centres of fasteners
Maximum edge distances
Back marks in channel flanges
Back marks in angles
Cross centres through flanges
Bolt capacities
Non-preloaded ordinary bolts in S275
Non-preloaded countersunk bolts in S275
Non-preloaded HSFG bolts in S275
Preloaded HSFG bolts in S275: non-slip in service
Preloaded HSFG bolts in S275: non-slip under factored loads
Preloaded countersunk HSFG bolts in S275: non-slip in service
Preloaded countersunk HSFG bolts in S275: non-slip under
factored loads
Non-preloaded ordinary bolts in S355
Non-preloaded countersunk bolts in S355
Non-preloaded HSFG bolts in S355
Preloaded HSFG bolts in S355: non-slip in service
Preloaded HSFG bolts in S355: non-slip under factored loads
1201
1202
1203
1204
1206
1207
1208
1214
1218
1220
1221
1222
1223
1224
1227
1228
1236
1236
1237
1237
1238
1240
1240
1241
1242
1244
1246
1247
1248
1249
1250
1251
1253
1255
1256
1257
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Contents
ix
Preloaded countersunk HSFG bolts in S355: non-slip in service
Preloaded countersunk HSFG bolts in S355: non-slip under factored
loads
Bolt and weld groups
Bolt group moduli – fasteners in the plane of the force
Bolt group moduli – fasteners not in the plane of the force
Weld group moduli – welds in the plane of the force
Capacities of fillet welds
Weld group moduli – welds not in the plane of the force
1258
1260
1264
1266
1270
1271
Other elements
Sheet pile sections
Larssen sections
Frodingham sections
Box sheet piles
High modulus piles
H-piles
Floor plate design tables
1274
1275
1276
1277
1279
1280
1259
Construction
Fire information sheets
Section factors for
universal beams
universal columns
circular hollow sections
rectangular hollow sections
rectangular hollow sections (square)
Minimum thickness of spray protection
Basic data on corrosion
1302
1303
1304
1305
1306
1307
1308
Codes and standards
British and European standards covering the design and construction of
steelwork
1311
Index
1323
1282
Full Page Ad, 3mm Bleed
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Introduction to sixth edition
At the instigation of the Iron and Steel Federation, the late Bernard Godfrey began
work in 1952 on the first edition of the Steel Designers’ Manual. As principal author
he worked on the manuscript almost continuously for a period of two years. On
many Friday evenings he would meet with his co-authors, Charles Gray, Lewis Kent
and W.E. Mitchell to review progress and resolve outstanding technical problems.
A remarkable book emerged. Within approximately 900 pages it was possible for
the steel designer to find everything necessary to carry out the detailed design of
most conventional steelwork. Although not intended as an analytical treatise, the
book contained the best summary of methods of analysis then available. The standard solutions, influence lines and formulae for frames could be used by the ingenious designer to disentangle the analysis of the most complex structure. Information
on element design was intermingled with guidance on the design of both overall
structures and connections. It was a book to dip into rather than read from cover
to cover. However well one thought one knew its contents, it was amazing how often
a further reading would give some useful insight into current problems. Readers
forgave its idiosyncrasies, especially in the order of presentation. How could anyone
justify slipping a detailed treatment of angle struts between a very general discussion of space frames and an overall presentation on engineering workshop design?
The book was very popular. It ran to four editions with numerous reprints in both
hard and soft covers. Special versions were also produced for overseas markets.
Each edition was updated by the introduction of new material from a variety of
sources. However, the book gradually lost the coherence of its original authorship
and it became clear in the 1980s that a more radical revision was required.
After 36 very successful years it was decided to rewrite and re-order the book,
while retaining its special character. This decision coincided with the formation of
the Steel Construction Institute and it was given the task of co-ordinating this
activity.
A complete restructuring of the book was undertaken for the fifth edition, with
more material on overall design and a new section on construction. The analytical
material was condensed because it is now widely available elsewhere, but all the
design data were retained in order to maintain the practical usefulness of the book
as a day-to-day design manual. Allowable stress design concepts were replaced by
limit state design encompassing BS 5950 for buildings and BS 5400 for bridges.
Design examples are to the more appropriate of these two codes for each particular application.
xi
Steel Designers' Manual - 6th Edition (2003)
xii
Introduction to sixth edition
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
The fifth edition was published in 1992 and proved to be a very worthy successor
to its antecedents. It also ran to several printings in both hard and soft covers; an
international edition was also printed and proved to be very popular in overseas
markets.
This sixth edition maintains the broad structure introduced in 1992, reflecting its
target readership of designers of structural steelwork of all kinds.
•
•
•
•
•
•
•
Design synthesis
Steel technology
Design theory
Element design
Connection design
Other elements
Construction.
Design synthesis: Chapters 1–5
A description of the nature of the process by which design solutions are arrived at
for a wide range of steel structures including:
•
•
•
•
Single- and multi-storey buildings (Chapters 1 and 2)
Heavy industrial frames (Chapter 3)
Bridges (Chapter 4)
Other diverse structures such as space frames, cable structures, towers and masts,
atria and steel in housing (Chapter 5).
Steel technology: Chapters 6–8
Background material sufficient to inform designers of the important problems
inherent in the production and use of steel, and methods of overcoming them in
practical design.
•
•
•
Applied metallurgy (Chapter 6)
Fatigue and Fracture (Chapter 7)
Sustainability and steel construction (Chapter 8).
Design theory: Chapters 9–12
A résumé of analytical methods for determining the forces and moments in structures subject to static or dynamic loads, both manual and computer-based.
Steel Designers' Manual - 6th Edition (2003)
Introduction to sixth edition
xiii
Comprehensive tables for a wide variety of beams and frames are given in the
Appendix.
•
•
•
•
Manual and computer analysis (Chapter 9)
Beam analysis (Chapter 10)
Frame analysis (Chapter 11)
Applicable dynamics (Chapter 12).
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Element design: Chapters 13–22
A comprehensive treatment of the design of steel elements, singly, in combination
or acting compositely with concrete.
•
•
•
•
•
•
•
•
•
•
Local buckling and cross-section classification (Chapter 13)
Tension members (Chapter 14)
Columns and struts (Chapter 15)
Beams (Chapter 16)
Plate girders (Chapter 17)
Members with compression and moments (Chapter 18)
Trusses (Chapter 19)
Composite floors (Chapter 20)
Composite beams (Chapter 21)
Composite columns (Chapter 22).
Connection design: Chapters 23–27
The general basis of design of connections is surveyed and amplified by consideration of specific connection methods.
•
•
•
•
•
Bolts (Chapter 23)
Welds and design for welding (Chapter 24)
Plate and stiffener elements in connections (Chapter 25)
Design of connections (Chapter 26)
Foundations and holding-down systems (Chapter 27).
Other elements: Chapters 28–30
•
•
•
Bearings and joints (Chapter 28)
Piles (Chapter 29)
Floors and orthotropic decks (Chapter 30).
Steel Designers' Manual - 6th Edition (2003)
xiv
Introduction to sixth edition
Construction: Chapters 31–35
Important aspects of steel construction about which a designer must be informed if
he is to produce structures which can be economically fabricated, and erected and
which will have a long and safe life.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
•
•
•
•
•
Tolerances (Chapter 31)
Fabrication (Chapter 32)
Erection (Chapter 33)
Fire protection and fire engineering (Chapter 34)
Corrosion resistance (Chapter 35).
Finally, Chapter 36 summarizes the state of progress on the Eurocodes, which will
begin to influence our design approaches from 2003 onwards.
A comprehensive collection of data of direct use to the practising designer is
compiled into a series of appendices.
By kind permission of the British Standards Institution, references are made to
British Standards throughout the manual. The tables of fabrication and erection
tolerances in Chapter 31 are taken from the National Structural Steelwork Specification, second edition. Much of the text and illustrations for Chapter 33 are taken
from Steelwork Erection by Harry Arch. Both these sources are used by kind
permission of the British Constructional Steelwork Association, the publishers.
These permissions are gratefully acknowledged.
Finally I would like to pay tribute both to the 38 authors who have contributed
to the sixth edition and to my hard-working principal editor, Dr Buick Davison. All
steelwork designers are indebted to their efforts in enabling this text book to be
maintained as the most important single source of information on steel design.
Graham Owens
Steel Designers' Manual - 6th Edition (2003)
Contributors
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Harry Arch
Harry Arch graduated from Manchester Faculty of Technology. For many years he
worked for Sir William Arrol, where he became a director, responsible for all outside
construction activities including major bridges, power stations and steelworks construction. In 1970 he joined Redpath Dorman Long International, working on offshore developments.
Mike Banfi
Mike Banfi joined Arup from Cambridge University in 1976. He has been involved
in the design of various major projects, including: Cummins Engine Plant, Shotts;
The Hong Kong and Shanghai Bank, Hong Kong; Usine L’Oreal, Paris; roofs for the
TGV stations, Lille and Roissy; roofs for the Rad-und Schwimmsportshalle, Berlin;
and various office blocks. He is now based in Arup Research & Development where
he provides advice on projects; examples include: Wellcome Wing to the Science
Museum, London; City Hall, London; T5, Heathrow. He is UK National Technical
Contact for Eurocode 4 part 1.1 and was on the steering committee for the 4th
edition of the NSSS. He is an Associate Director.
Hubert Barber
Hubert Barber joined Redpath Brown in 1948 and for five years gained a wide
experience in steel construction. The remainder of his working life was spent in local
government, first at Manchester and then in Yorkshire where he became chief structural engineer of West Yorkshire. He also lectured part-time for fourteen years at
the University of Bradford.
Tony Biddle
Tony Biddle graduated in civil engineering from City University in 1966 and spent
the early part of his career in contractors, designing in steel and reinforced concrete
xv
Steel Designers' Manual - 6th Edition (2003)
xvi
Contributors
before specializing in soil mechanics and foundation design in 1970. Between 1974
and 1993 he worked in the offshore industry, becoming a specialist in steel piling.
He joined SCl in 1994 as manager for civil engineering and has developed the R&D
research project work in steel piling related topics. He has been a drafting member
for Eurocode 3 part 5, contributor to BS 8002 amendments, and author of several
SCl publications.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Michael Burdekin
Michael Burdekin graduated from Cambridge University in 1959.After fifteen years
of industrial research and design experience he went to UMIST, where he is now
Professor of Civil and Structural Engineering. His specific expertise is the field of
welded steel structures, particularly in the application of fracture mechanics to fracture and fatigue failure.
Brian Cheal
Brian Cheal graduated from Brighton Technical College in 1951 with an External
Degree of the University of London. He was employed with W.S. Atkins and Partners from 1951 to 1986, becoming a technical director in 1979, and specialized in the
analysis and design of steel-framed structures, including heavy structural framing
for power stations and steelworks. He has written design guides and given lectures
on various aspects of connection design and is co-author of Structural Steelwork
Connections.
David Dibb-Fuller
David Dibb-Fuller started his career with the Cleveland Bridge and Engineering
Company in London. His early bridge related work gave a strong emphasis to heavy
fabrication; in later years he moved on to building structures. As technical director
for Conder Southern in Winchester his strategy was to develop close links between
design for strength and design for production. Currently he is a partner with Gifford
and Partners in Southampton where he continues to exercise his skills in the design
of steel structures.
Ian Duncan
Ian Duncan joined the London office of Ove Arup and Partners in 1966 after
graduating from Surrey University. From 1975 he taught for four years at Univer-
Steel Designers' Manual - 6th Edition (2003)
Contributors
xvii
sity College Cardiff before joining Buro Happold. He now runs his own practice in
Bristol.
Michael Green
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Michael Green graduated from Liverpool University in 1971. After an early career
in general civil engineering, he joined Buro Happold, where he is now an executive
partner. He has worked on a wide variety of building projects, developing a specialist expertise in atria and large-span structures.
Alan Hart
Alan Hart graduated from the University of Newcastle upon Tyne in 1968 and
joined Ove Arup and Partners. During his career he has been involved in the design
of a number of major award-winning buildings, including Carlsberg Brewery,
Northampton; Cummins Engine Plant, Shotts, Lanarkshire; and the Hongkong and
Shanghai Bank, Hong Kong. He is a project director of Ove Arup and Partners.
Alan Hayward
Alan Hayward is a bridge specialist and is principal consultant of Cass Hayward
and Partners, who design and devise the erection methodology for all kinds of steel
bridges, many built on a design : construct basis. Projects include London Docklands
Light Railway viaducts, the M25/M4 interchange, the Centenary Lift bridge at Trafford Park and the Newark Dyke rail bridge reconstruction. Movable bridges and
roll-on/roll-off linkspans are also a speciality. He is a former chief examiner for the
Institution of Structural Engineers and was invited to become a Fellow of the Royal
Academy of Engineers in 2001.
Eric Hindhaugh
Eric Hindhaugh trained as a structural engineer in design and constructional steelwork, timber and lightweight roll-formed sections. He then branched into promotional and marketing activities. He was a market development manager in
construction for British Steel Strip Products, where he was involved in Colorcoat
and the widening use of lightweight steel sections for structural steel products. He
is now retired.
Steel Designers' Manual - 6th Edition (2003)
xviii
Contributors
Roger Hudson
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Roger Hudson studied metallurgy at Sheffield Polytechnic whilst employed by
BISRA. He also has a Masters degree from the University of Sheffield. In 1968, he
joined the United Steel Companies at Swinden Laboratories in Rotherham to work
on the corrosion of stainless steels. The laboratories later became part of British
Steel where he was responsible for the Corrosion Laboratory and several research
projects. He is now principal technologist in the recently formed Corus company.
He is a member of several technical and international standards committees, has
written technical publications, and has lectured widely on the corrosion and
protection of steel in structures. He is a long serving professional member of the
Institute of Corrosion and is currently chairman of the Yorkshire branch and chairman of the Training and Certification Governing Board.
Ken Johnson
Ken Johnson was head of corrosion and coatings at British Steel’s Swinden
Laboratories. His early experience was in the paint industry but he then worked in
steel for over twenty-five years, dealing with the corrosion and protection aspects
of the whole range of British Steel’s products, including plates, section, piling, strip
products, tubes, stainless steels, etc. He represented the steel industry on several BSI
and European Committees and was a council member of the Paint Research
Association. He is now retired.
Alan Kwan
Alan Kwan graduated from the University of Sheffield and Cambridge University.
He is a lecturer in structural engineering at Cardiff University, specializing in lightweight, deployable, tension and space structures, and numerical methods for their
analysis.
Mark Lawson
A graduate of Imperial College, and the University of Salford, where he worked in
the field of cold-formed steel, Mark Lawson spent his early career at Ove Arup and
Partners and the Construction Industry Research and Information Association. In
1987 he joined the newly formed Steel Construction Institute as research manager
for steel in buildings, with particular reference to composite construction, fire engineering and cold-formed steel. He is a member of the Eurocode 4 project team on
fire-resistant design.
Steel Designers' Manual - 6th Edition (2003)
Contributors
xix
Ian Liddell
After leaving Cambridge, Ian Liddell joined Ove Arup and Partners to work on the
roof of the Sydney Opera House and on the South Bank Art Centre. His early career
encompassed a wide range of projects, with particular emphasis on shell structures
and lightweight tension and fabric structures. Since 1976 he has been a partner of
Buro Happold and has been responsible for a wide range of projects, many with
special structural engineering features, including mosques, auditoriums, mobile and
temporary structures, stadiums and retail atria.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Matthew Lovell
Matthew Lovell studied civil engineering at University College, London. After
graduation Matthew worked for Arup on the Chur Station roof project. He is now
senior associate at Buro Happold and has worked on many steel structures, including Thames Valley University LRC, the National Centre for Popular Music, and St
David’s RF Hotel. He has recently completed an MSc in Interdisciplinary Design
at Cambridge University.
Stephen Matthews
Stephen Matthews graduated from the University of Nottingham in 1974 and completed postgraduate studies at Imperial College in 1976–77. His early professional
experience was gained with Rendel Palmer and Tritton. During subsequent employment with Fairfield Mabey and Cass Hayward and Partners he worked on the design
of several large composite bridges, including the Simon de Montfort Bridge
Evesham, M25/M4 interchange, Poyle, and viaducts on the Docklands Light
Railway. He is a director of WSP (Civils), where he has been manager of the Bridges
Division since 1990. Work has included a number of major bridge repair schemes
and drafting of the UK National Application Document for Eurocode 3 part 2
(steel bridges).
David Moore
David Moore graduated from the University of Bradford in 1981 and joined the
Building Research Establishment (BRE) where he has completed over twenty years
of research and specialist advisory work in the area of structural steelwork. He is
the author of over 70 technical papers on a wide range of subjects. He has also made
a significant contribution to a number of specialist steel and composite connection
design guides, many of which are used daily by practising structural engineers and
Steel Designers' Manual - 6th Edition (2003)
xx
Contributors
steelwork fabricators. Currently he is the director of the Centre for Structural Engineering at BRE.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Rangachari Narayanan
Rangachari Narayanan graduated in civil engineering from Annamalai University
(India) in 1951. In a varied professional career spanning over forty years, he has
held senior academic positions at the Universities of Delhi, Manchester and Cardiff.
He is the recipient of several awards including the Benjamin Baker Gold Medal
and George Stephenson Gold Medal, both from the Institution of Civil Engineers.
For many years he headed the Education and Publication Divisions at the Steel
Construction Institute.
David Nethercot
Since graduating from the University of Wales, Cardiff, David Nethercot has completed thirty years of teaching, research and specialist advisory work in the area of
structural steelwork. The author of over 300 technical papers, he has lectured frequently on post-experience courses; he is chairman of the BSI Committee responsible for BS 5950, and is a frequent contributor to technical initiatives associated
with the structural steelwork industry. Since 1999 he has been head of the Department of Civil and Environmental Engineering at Imperial College.
Gerard Parke
Gerry Parke is a lecturer in structural engineering at the University of Surrey
specializing in the analysis and design of steel structures. His particular interests
lie in assessing the collapse behaviour of both steel industrial buildings and largespan steel space structures.
Phil Peacock
Phil Peacock is a member of the Corus Construction Centre. He started his career
in 1965 at steelwork fabricators Ward Bros. Ltd., gained an HND at Teesside Polytechnic and moved to White Young Consulting Engineers in 1973 before joining
British Steel (now Corus) in 1988. His experience covers the design management
of a wide range of projects: heavy plant buildings and structures for the steel, petrochemical and coal industries, commercial offices, leisure and retail developments.
He serves on several industry committees and is a past chairman of the Institution
of Structural Engineers Scottish Branch.
Steel Designers' Manual - 6th Edition (2003)
Contributors
xxi
Alan Pottage
Alan Pottage graduated from the University of Newcastle upon Tyne in 1976 and
gained a Masters degree in structural steel design from Imperial College in 1984.
He has gained experience in all forms of steel construction, particularly portal frame
and multi-rise structures, and has contributed to various code committees, and SCI
guides on composite design and connections.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Graham Raven
Graham Raven graduated from King’s College, London in 1963 and joined Ove
Arup and Partners. Following thirteen years with consulting engineers working on
a variety of building structures he joined a software house pioneering work in structural steel design and detailing systems. In 1980 this experience took him to Ward
Building Systems where he became technical director and was closely associated
with the development of a range of building components and increased use
of welded sections in buildings. Since 1991, with the exception of a year with a
software house specialising in 3D detailing systems, he has been employed at the
Steel Construction Institute, where he is the senior manager responsible for the
Sustainability Group.
John Righiniotis
John Righiniotis graduated from the University of Thessalonika in 1987 and
obtained an MSc in structural steel design from Imperial College in 1988. He
worked at the Steel Construction Institute on a wide range of projects until June
1990 when he was required to return to Greece to carry out his military service.
John Roberts
John Roberts graduated from the University of Sheffield in 1969 and was awarded
a PhD there in 1972 for research on the impact loading of steel structures. His professional career includes a period of site work with Alfred McAlpine, following
which he has worked as a consulting engineer, since 1981 with Allott &
Lomax/Babtie Group. He is a director of Babtie Group where he heads up the
Structures and Buildings Teams that have designed many major steelwork structures. He was president of the Institution of Structural Engineers in 1999–2000 and
is a council member of both the Steel Construction Institute and the BCSA.
Steel Designers' Manual - 6th Edition (2003)
xxii
Contributors
Terry Roberts
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Terry Roberts graduated in civil and structural engineering from the University of
Wales Cardiff in 1967, and following three years of postgraduate study was awarded
a PhD in 1971. His early professional experience was gained in bridge design and
site investigation for several sections of the M4 motorway in Wales. He returned to
academic life in 1975. He is the author of over 100 technical papers on various
aspects of structural engineering, for which he received a DSc from the University
of Wales and the Moisseiff Award from the Structural Engineering Institute of the
American Society of Civil Engineers in 1997. Since 1996 he has been head of the
Division of Structural Engineering in the Cardiff School of Engineering.
Jef Robinson
Jef Robinson graduated in metallurgy from Durham University in 1962. His early
career in the steel industry included formulating high ductility steels for automotive applications and high-strength notch ductile steels for super tankers, drilling
platforms and bridges. Later as market development manager for the structural division of British Steel (now Corus) he chaired the BSI committee that formulated BS
5950 Part 8: Fire Resistant Design for structural steelwork and served on a number
of international fire related committees. He was appointed honorary professor at
the University of Sheffield in 2000.
Alan Rogan
Alan Rogan is a leading consultant to the steel industry, working with prestigious
clients such as Corus and Cleveland Bridge Engineering Group. Alan has been
involved in the construction of many buildings, such as Canary Wharf, Gatwick
Airport extension and many bridges from simple footbridges to complex multispans, in the UK and overseas.
Dick Stainsby
Dick Stainsby’s career training started with an HNC and went on to include postgraduate studies at Imperial College London. His experience has encompassed steel
structures of all kinds including bridgework. He was for many years chief designer
with Redpath Dorman Long Middlesbrough. Since retiring from mainstream industry he has assisted the British Constructional Steelwork Association, the Steel Construction Institute and the Institution of Structural Engineers in the production of
Steel Designers' Manual - 6th Edition (2003)
Contributors
xxiii
technical publications relating to steelwork connections. He also compiled the
National Structural Steelwork Specification for Building Construction, which is now
in its 4th Edition.
Paul Tasou
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Paul Tasou graduated from Queen Mary College, London in 1978 and subsequently
obtained an MSc in structural steel design from Imperial College, London. He
spent eleven years at Rendel Palmer and Tritton working on a wide range of bridge,
building and civil engineering projects. He is now principal partner in Tasou
Associates.
Colin Taylor
Colin Taylor graduated from Cambridge in 1959. He started his professional career
in steel fabrication, initially in the West Midlands and subsequently in South India.
After eleven years he moved into consultancy where, besides practical design, he
became involved with graduate training, the use of computers for design and drafting, company technical standards and drafting work for British Standards and for
Eurocodes as editorial secretary for Eurocode 3. Moving to the Steel Construction
Institute on its formation as manager of the Codes and Advisory Division, he also
became involved with the European standard for steel fabrication and erection
Execution of Steel Structures.
John Tyrrell
John Tyrrell graduated from Aston University in 1965 and immediately joined Ove
Arup and Partners. He has worked for them on a variety of projects in the UK,
Australia and West Africa; he is now a project director. He has been responsible for
the design of a wide range of towers and guyed masts. He currently leads the Industrial Structures Group covering diverse fields of engineering from telecommunications and broadcasting to the power industry.
Peter Wickens
Having graduated from Nottingham University in 1971, Peter Wickens spent much
of his early career in civil engineering, designing bridges and Metro stations. In 1980,
Steel Designers' Manual - 6th Edition (2003)
xxiv
Contributors
he changed to the building structures field and was project engineer for the Billingsgate Development, one of the first of the new generation of steel composite buildings. He is currently manager of the Structural Division and head of discipline for
Building Structures at Mott MacDonald.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Michael Willford
Michael Willford joined Arup in 1975, having graduated from Cambridge University. He has been a specialist in the design of structures subjected to dynamic actions
for over twenty years. His design and analysis experience covers a wide variety of
projects including buildings, bridges and offshore structures. He is currently a director of Arup and the leader of a team of specialists working in these fields based in
London and San Francisco.
John Yates
John Yates was appointed to a personal chair in mechanical engineering at the
University of Sheffield in 2000 after five years as a reader in the department. He
graduated from Pembroke College, Cambridge in 1981 in metallurgy and materials
science and then undertook research degrees at Cranfield and the University of
Sheffield before several years of postdoctoral engineering and materials research.
His particular interests are in developing structural integrity assessment tools based
on the physical mechanisms of fatigue and fracture. He is the honorary editor
of Engineering Integrity and an editor of the international journal Fatigue and
Fracture of Engineering Materials and Structures.
Ralph Yeo
Ralph Yeo graduated in metallurgy at Cardiff and Birmingham and lectured at The
University of the Witwatersrand. In the USA he worked on the development of
weldable high-strength and alloy steels with International Nickel and US Steel and
on industrial gases and the development of welding consumables and processes at
Union Carbide’s Linde Division. Commercial and general management activities in
the UK, mainly with The Lincoln Electric Company, were followed by twelve years
as a consultant and expert witness, with special interest in improved designs for
welding.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Notation
Several different notations are adopted in steel design; different specializations frequently give different meanings to the same symbol. These differences have been
maintained in this book. To do otherwise would be to separate this text both from
other literature on a particular subject and from common practice. The principal
definitions for symbols are given below. For conciseness, only the most commonly
adopted subscripts are given; others are defined adjacent to their usage.
A
or
or
Ae
Ag
As
At
Av
a
or
or
or
or
B
B
b
or
or
be
b1
C
C
Cv
Cy
c
or
Area
End area of pile
Constant in fatigue equations
Effective area
Gross area
Shear area of a bolt
Tensile stress area of a bolt
Shear area of a section
Spacing of transverse stiffeners
Effective throat size of weld
Crack depth
Distance from central line of bolt to edge of plate
Shaft area of pile
Breadth
Transformation matrix
Outstand
Width of panel
Distance from centreline of bolt to toe of fillet weld or to half of root
radius as appropriate
Effective breadth or effective width
Stiff bearing length
Crack growth constant
Transformation matrix
Charpy impact value
Damping coefficient
Bolt cross-centres
Cohesion of clay soils
xxv
Steel Designers' Manual - 6th Edition (2003)
xxvi
Notation
D
D
Dr
Ds
d
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
or
or
de
E
e
ey
Fc
Fs
Ft
Fv
f
fa
fc
fcu
fm
fr
fv
or
G
g
H
or
h
or
or
Io
Ioo
Ix
Ixx
Iy
Iyy
K
or
K
Ks
or
K1, K2, K3
Depth of section
Diameter of section or hole
Elasticity matrix
Profile height for metal deck
Slab depth
Depth of web
Nominal diameter of fastener
Depth
Effective depth of slab
Modulus of elasticity of steel (Young’s modulus)
End distance
Material yield strain
Compressive force due to axial load
Shear force (bolts)
Tensile force
Shear force (sections)
Flexibility coefficient
Longitudinal stress in flange
Compressive stress due to axial load
Cube strength of concrete
Force per unit length on weld group from moment
Resultant force per unit length on weld group from applied concentric
load
Force per unit length on weld group from shear
Shear stress
Shear modulus of steel
Gravitational acceleration
Warping constant of section
Horizontal reaction
Height
Stud height
Depth of overburden
Polar second moment of area of bolt group
Polar second moment of area of weld group of unit throat about polar
axis
Second moment of area about major axis
Polar second moment of area of weld group of unit throat about xx axis
Second moment of area about minor axis
Polar second moment of area of weld group of unit throat about yy axis
Degree of shear connection
Stiffness
Stiffness matrix
Curvature of composite section from shrinkage
Constant in determining slip resistance of HSFG bolts
Empirical constants defining the strength of composite columns
Steel Designers' Manual - 6th Edition (2003)
Notation
ka
kd
kp
L
Ly
M
or
Max, May
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Mb
ME
Mo
Mpc
Mrx, Mry
Mx, My
Mx , My
M1, M2
m
or
or
md
N
Nc, Nq, Ng
n
or
or
or
P
or
or
Pbb
Pbg
Pbs
Pc
Pcx, Pcy
Po
Ps
PsL
Pt
Pu
Pv
xxvii
Coefficient of active pressure
Empirical constant in composite slab design
Coefficient of passive resistance
Length of span or cable
Shear span length of composite slab
Moment
Larger end moment
Maximum buckling moment about major or minor axis in presence
of axial load
Buckling resistance moment (lateral – torsional)
Elastic critical moment
Mid-length moment on a simply-supported span equal to unrestrained
length
Plastic moment capacity of composite section
Reduced moment capacity of section about major or minor axis in
the presence of axial load
Applied moment about major or minor axis
Equivalent uniform moment about major or minor axis
End moments for a span of a continuous composite beam
Equivalent uniform moment factor
Empirical constant in fatigue equation
Number of vertical rows of bolts
Empirical constant in composite slab design
Number of cycles to failure
Constants in Terzaghi’s equation for the bearing resistance of clay
soils
Crack growth constant
Number of shear studs per trough in metal deck
Number of horizontal rows of bolts
Distance from bolt centreline to plate edge
Force in structural analysis
Load per unit surface area on cable net
Crushing resistance of web
Bearing capacity of a bolt
Bearing capacity of parts connected by friction-grip fasteners
Bearing capacity of parts connected by ordinary bolts
Compression resistance
Compression resistance considering buckling about major or minor axis
only
Minimum shank tension for preloaded bolt
Shear capacity of a bolt
Slip resistance provided by a friction-grip fastener
Tension capacity of a member or fastener
Compressive strength of stocky composite column
Shear capacity of a section
Steel Designers' Manual - 6th Edition (2003)
xxviii
Notation
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
p
pb
pbb
pbg
pbs
pc
pE
po
ps
pt
pw
py
Q
q
qb
qcr
qe
qf
R
or
or
Rc
Rq
Rr
Rs
Rw
r
rr
rx, ry
S
SR
Sx, Sy
s
or
T
or
t
U
Us
u
V
or
Vb
Ratio of cross-sectional area of profile to that of concrete in a composite slab
Bending strength
Bearing strength of a bolt
Bearing strength of parts connected by friction-grip fasteners
Bearing strength of parts connected by ordinary bolts
Compressive strength
Euler strength
Minimum proof stress of a bolt
Shear strength of a bolt
Tension strength of a bolt
Design strength of a fillet weld
Design strength of steel
Prying force
Ultimate bearing capacity
Basic shear strength of a web panel
Critical shear strength of a web panel
Elastic critical shear strength of a web panel
Flange-dependent shear strength factor
Reaction
Load applied to bolt group
Radius of curvature
Compressive capacity of concrete section in composite construction
Capacity of shear connectors between point of contraflexure and point
of maximum negative moment in composite construction
Tensile capacity in reinforcement in composite construction
Tensile capacity in steel section in composite construction
Compression in web section in composite construction
Root radius in rolled section
Reduction factor in composite construction
Radius of gyration of a member about its major or minor axis
Span of cable
Applied stress range
Plastic modulus about major or minor axis
Spacing
Leg length of a fillet weld
Thickness of a flange or leg
Tension in cable
Thickness of web
Elastic energy
Specified minimum ultimate tensile strength of steel
Buckling parameter of the section
Shear force
Shear resistance per unit length of beam in composite construction
Shear buckling resistance of stiffened web utilizing tension field action
Steel Designers' Manual - 6th Edition (2003)
Notation
Vcr
v
W
or
or
or
w
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
or
or
Xe
x
xp
Y
Ys
Zc
Zoo
Zx, Zy
z
a
ae
b
or
g
or
or
gf
gm
D
d
or
dc
dic
do
doo
e
or
h
l
lcr
lLO
lLT
xxix
Shear buckling resistance of stiffened or unstiffened web without utilizing tension field action
Slenderness factor for beam
Point load
Foundation mass
Load per unit length on a cable
Energy required for crack growth
Lateral displacement
Effective width of flange per bolt
Uniformly distributed load on plate
Elastic neutral axis depth in composite section
Torsion index of section
Plastic neutral axis depth in composite section
Correction factor in fracture mechanics
Specified minimum yield stress of steel
Elastic section modulus for compression
Elastic modulus for weld group of unit throat subject to torsional load
Elastic modulus about major or minor axis
Depth of foundation
Coefficient of linear thermal expansion
Modular ratio
Ratio of smaller to larger end moment
Coefficient in determination of prying force
Ratio M/Mo, i.e. ratio of larger end moment to mid-length moment on
simply-supported span equal to unrestrained length
Bulk density of soil
Coefficient in determination of prying force
Overall load factor
Material strength factor
Displacements in vector
Deflection
Elongation
Deflection of composite beam at serviceability limit state
Deflection of composite beam at serviceability limit state in
presence of partial shear connection
Deflection of steel beam at serviceability limit state
Deflection in continuous composite beam at serviceability limit state
Constant (275/py)1/2
Strain
Load ratio for composite columns
Slenderness, i.e. effective length divided by radius of gyration
Elastic critical load factor
Limiting equivalent slenderness
Equivalent slenderness
Steel Designers' Manual - 6th Edition (2003)
xxx
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
lo
m
mx, my
s
se
f
Notation
Limiting slenderness
Slip factor
Reduction factors on moment resistance considering axial load
Stress
Tensile stress
Diameter of composite column
or Angle of friction in granular soil
Steel Designers' Manual - 6th Edition (2003)
Chapter 1
Single-storey buildings
by GRAHAM RAVEN and ALAN POTTAGE
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
1.1 Range of building types
It is estimated that around 50% of the hot-rolled constructional steel used in
the UK is fabricated into single-storey buildings, being some 40% of the total steel
used in them. The remainder is light-gauge steel cold-formed into purlins, rails.
cladding and accessories. Over 90% of single-storey non-domestic buildings have
steel frames, demonstrating the dominance of steel construction for this class of
building. These relatively light, long-span, durable structures are simply and quickly
erected, and developments in steel cladding have enabled architects to design
economical buildings of attractive appearance to suit a wide range of applications
and budgets.
The traditional image was a dingy industrial shed, with a few exceptions such as
aircraft hangars and exhibition halls. Changes in retailing and the replacement of
traditional heavy industry with electronics-based products have led to a demand for
increased architectural interest and enhancement.
Clients expect their buildings to have the potential for easy change of layout
several times during the building’s life. This is true for both institutional investors
and owner users. The primary feature is therefore flexibility of planning, which, in
general terms, means as few columns as possible consistent with economy. The
ability to provide spans up to 60 m, but most commonly around 30 m, gives an
extremely popular structural form for the supermarkets, do-it-yourself stores and
the like which are now surrounding towns in the UK. The development of steel
cladding in a wide variety of colours and shapes has enabled distinctive and attractive forms and house styles to be created.
Improved reliability of steel-intensive roofing systems has contributed to their
acceptability in buildings used by the public and perhaps more importantly in ‘hightech’ buildings requiring controlled environments. The structural form will vary
according to span, aesthetics, integration with services, cost and suitability for the
proposed activity. A cement manufacturing building will clearly have different
requirements from a warehouse, food processing plant or computer factory.
The growth of the leisure industry has provided a challenge to designers, and
buildings vary from the straightforward requirement of cover for bowls, tennis, etc.,
to an exciting environment which encourages people to spend days of their holidays indoors at water centres and similar controlled environments suitable for year
round recreation.
1
Steel Designers' Manual - 6th Edition (2003)
Single-storey buildings
2
40
35
unit weight
(kg/rn2
30
of floor
area)
25
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
20
15
10
15
20
25
30
35
40
45
span (metres)
Fig. 1.1
Comparison of bare frame weights for portal and lattice structures
In all instances the requirement is to provide a covering to allow a particular activity to take place; the column spacing is selected to give as much freedom of use of
the space consistent with economy. The normal span range will be from 12 m to
50 m, but larger spans are feasible for hangars and enclosed sports stadia.
Figure 1.1 shows how steel weight varies with structural form and span.1
1.2 Anatomy of structure
A typical single-storey building consisting of cladding, secondary steel and a frame
structure is shown in Fig. 1.2.
1.2.1 Cladding
Cladding is required to be weathertight, to provide insulation, to have penetrations
for daylight and access, to be aesthetically pleasing, and to last the maximum time
with a minimum of maintenance consistent with the budget.
The requirements for the cladding to roofs and walls are somewhat different.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Anatomy of structure
Fig. 1.2
3
Structural form for portal-frame building (some rafter bracing omitted for clarity)
The ability of the roof to remain weathertight is clearly of paramount importance,
particularly as the demand for lower roof pitches increases, whereas aesthetic
considerations tend to dictate the choice of walling.
Over the past 30 years, metal cladding has been the most popular choice for both
roofs and walls, comprising a substrate of either steel or aluminium.
Cladding with a steel substrate tends to be more economical from a purely
cost point of view and, coupled with a much lower coefficient of thermal expansion
than its aluminium counterpart, has practical advantages. However, the integrity of
the steel substrate is very much dependent on its coatings to maintain resistance to
corrosion. In some ‘sensitive’ cases, the use of aluminium has been deemed to
better serve the specification.
Typical external and internal coatings for steel substrates manufactured by Corus
(formerly British Steel plc/Hoogovens) are detailed below.
Substrate – steel
•
•
Galvatite, hot-dipped zinc coated steel to BS EN10147: 1992. Grade Fe E220G
with a Z275 zinc coating.
Galvalloy, hot-dipped alloy-coated steel substrate (95% zinc, 5% aluminium)
to BS EN 10214. Grade S220 GD+ZA with a ZA255 alloy coating.
Steel Designers' Manual - 6th Edition (2003)
Single-storey buildings
4
Coatings – external
•
•
•
Colorcoat HPS200 – 200 mm coating applied to the weatherside of the sheet on
Galvalloy, above. Provides superior durability, colour stability and corrosion
resistance.
Colorcoat PVF2 – 27 mm, stoved fluorocarbon, coating on Galvatite, above.
Provides excellent colour stability.
Colorcoat Silicon Polyester – An economic coating on Galvatite, above. Provides
medium term durability for worldwide use.
Coatings – internal
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
•
•
•
Colorcoat Lining Enamel – 22 mm coating, ‘bright white’, with an easily cleaned
surface.
Colorcoat HPS200 Plastisol – 200 mm coating, used in either a corrosive
environment or one of high internal humidity.
Colorcoat Stelvetite Foodsafe – 150 mm coating, comprising a chemically inert
polymer for use in cold stores and food processing applications.
The reader should note that there is an increasing move towards whole life-cycle
costing of buildings in general, on which the cladding element has a significant influence. A cheaper cladding solution at the outset of a project may result in a smaller
initial outlay for the building owner. Over the life of the building, however, running
costs could offset (and possibly negate) any savings that may have accrued at
procurement stage. A higher cladding specification will reduce not only heating
costs but also carbon dioxide (CO2) emissions.
The construction of the external skin of a building can take several forms, the
most prevalent being:
(1)
(2)
(3)
(4)
single-skin trapezoidal
double-skin trapezoidal shell
standing seam with concealed fixings
composite panels.
Further information on the above topics can be found by reference to the cladding
manufacturers’ technical literature, and section 1.4.8 below.
1.2.2 Secondary elements
In the normal single-storey building the cladding is supported on secondary
members which transmit the loads back to main structural steel frames. The spacing
of the frames, determined by the overall economy of the building, is normally in the
range 5–8 m, with 6 m and 7.5 m as the most common spacings.
Steel Designers' Manual - 6th Edition (2003)
Anatomy of structure
5
120<d<300
Zed
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 1.3
modified Zed
Sigma
Popular purlin and frame sections
A combination of cladding performance, erectability and the restraint requirements for economically-designed main frames dictates that the purlin and rail
spacing should be 1.5–2 m.
For this range the most economic solution has proved to be cold-formed lightgauge sections of proprietary shape and volume produced to order on computer
numerically controlled (CNC) rolling machines. These have proved to be extremely
efficient since the components are delivered to site pre-engineered to the exact
requirements. which minimizes fabrication and erection times and eliminates material wastage. Because of the high volumes, manufacturers have been encouraged to
develop and test all material-efficient sections. These fall into three main categories:
Zed, modified Zed and Sigma sections. Figure 1.3 illustrates the range.
The Zed section was the first shape to be introduced. It is material-efficient but
the major disadvantage is that the principal axes are inclined to the web. If subject
to unrestrained bending in the plane of the web, out-of-plane displacements occur:
if these are restrained, out-of-plane forces are generated.
More complicated shapes have to be rolled rather than press braked. This is a
feature of the UK, where the market is supplied by relatively few manufacturers
and the volumes produced by each allow the advanced manufacturing techniques
to be employed, giving competitive products and service.
As roof pitches become lower, modified Zed sections have been developed
with the inclination of the principal axis considerably reduced, so enhancing overall
performance. Stiffening has been introduced, improving material efficiency.
The Sigma shape, in which the shear centre is approximately coincident with the
load application line, has advantages. One manufacturer now produces, using rolling,
a third-generation product of this configuration, which is economical.
1.2.3 Primary frames
The frame supports the cladding but, with increasing architectural and service
demands, other factors are important. The basic structural form has developed
against the background of achieving the lowest cost envelope by enclosing the
minimum volume. Plastic design of portal frames brings limitations on the spacing
of restraints of around 1.8–2 m. The cladding profiles are economic in this range:
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
6
Single-storey buildings
they can support local loads and satisfy drainage requirements. The regime is therefore for the loads to be transferred from the sheeting on to the purlins and rails,
which in turn must be supported on a primary structure. Figure 1.4 shows the simplest possible type of structure with vertical columns and a horizontal spanning
beam. There is a need for a fall in the roof finish to provide drainage, but for small
spans the beam can be effectively horizontal with the fall being created in the
finishes or by a nominal slope in the beam. The minimum slope is also a function
of weatherproofing requirements of the roof material.
The simple form shown would be a mechanism unless restraint to horizontal
forces is provided. This is achieved either by the addition of bracing in both plan
and vertical planes or by the provision of redundancies in the form of momentresisting joints. The important point is that all loads must be transmitted to the foundations in a coherent fashion even in the simplest of buildings, whatever their size.
The range of frame forms is discussed in more detail in later sections but Fig. 1.5
shows the structural solutions commonly used.The most common is the portal shape
with pinned bases, although this gives a slightly heavier frame than the fixed-base
option. The overall economy, including foundations, is favourable. The portal form
is both functional and economic with overall stability being derived from the
provision of moment-resisting connections at eaves and apex.
The falls required to the roof are provided naturally with the cladding being
carried on purlins, which in turn are supported by the main frame members.
Architectural pressures have led to the use of flatter slopes compatible with
weathertightness; the most common is around 6°, but slopes as low as 1° are used,
which means deflection control is increasingly important.
Traditionally, portal frames have been fabricated from compact rolled sections
and designed plastically. More recently the adoption of automated welding techniques has led to the introduction of welded tapered frames, which have been extensively used for many years in the USA. For economy these frames have deep slender
sections and are designed elastically. In addition to material economies, the benefit
is in the additional stiffness and reduced deflections.
Although the portal form is inherently pleasing to the eye, given a well-proportioned and detailed design, the industrial connotation, together with increased
Fig. 1.4
Simplest single-storey structure
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Anatomy of structure
Fig. 1.5
7
A range of structural forms
service requirements, has encouraged the use of lattice trusses for the roof structure. They are used both in the simple forms with fixed column bases and as portal
frames with moment-resisting connections between the tops of the columns for
long-span structures such as aircraft hangars, exhibition halls and enclosed sports
facilities.
1.2.4 Resistance to sway forces
Most of the common forms provide resistance to sidesway forces in the plane of the
frame. It is essential also to provide resistance to out-of-plane forces; these are
usually transmitted to the foundations with a combination of horizontal and vertical girders. The horizontal girder in the plane of the roof can be of two forms
as shown in Fig. 1.6. Type (a) is formed from members, often tubes, capable of
carrying tension or compression. One of the benefits is in the erection stage as the
braced bay can be erected first and lined and levelled to provide a square and safe
springboard for the erection of the remainder.
Steel Designers' Manual - 6th Edition (2003)
8
Single-storey buildings
compression
strut
tension
members
(0)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 1.6
(b)
Roof bracing
Type (b) uses less material but more members are required. The diagonals are
tension-only members (wire rope has been used) and the compression is taken
in the orthogonal strut which has the shortest possible effective length. It may be
possible to use the purlins, strengthened where necessary, for this purpose.
Similar arrangements must be used in the wall to carry the forces down to
foundation level. If the horizontal and vertical girders are not in the same bay, care
must be taken to provide suitable connecting members.
1.3 Loading
1.3.1 External gravity loads
The dominant gravity loading is snow, with a general case being the application of
a minimum basic uniform snow load of 0.60 kN/m2 as an imposed load. However,
there are certain geographical areas where this basic minimum will be unduly
conservative, a fact that is recognised in BS 6399: Part 3.2
BS 6399: Part 3 contains a map showing values of ‘basic snow load on the ground’
in a form similar to a ‘contour map’. In ascertaining the snow load for which the
structure is to be designed, the designer must take cognisance of the latter value at
the building’s location and the altitude of the building above sea level. By substituting the latter values in a simple algorithm, the site snow load can be determined
and used in building design. (It should be noted that any increase in the value of
snow load on the ground only takes effect at altitudes greater than 100 m.)
The trend towards both curved and multi-span pitched and curved roof
structures, with eaves and gable parapets, further adds to the number of load
combinations that the designer must recognise.
BS 6399: Part 3 recognizes the possibility of drifting in the valleys of multi-span
structures and adjacent to parapets, in addition to drifting at positions of abrupt
changes in height. The process that must be followed by the designer in order to
arrive at the relevant load case is illustrated by means of diagrams and associated
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Loading
9
flow charts. In all instances, the drift is idealized as a varying, triangularly
distributed load.
The drift condition must be allowed for not only in the design of the frame
itself, but also in the design of the purlins that support the roof cladding, since the
intensity of loading at the position of maximum drift is often far in excess of the
minimum basic uniform snow load.
In practice, the designer will invariably design the purlins for the uniform load
case, thereby arriving at a specific section depth and gauge. In the areas subject to
drift, the designer will maintain that section and gauge by reducing the purlin
spacing local to the greater loading in the area of maximum drift. (In some instances,
however, it may be possible to maintain purlin depth but increase purlin gauge in
the area of the drift. An increase in purlin gauge implies a stronger purlin, which in
turn implies that the spacing of the purlins may be increased over that of a thinner
gauge. However, there is the possibility on site that purlins which appear identical
to the eye, but are of different gauge, may not be positioned in the location that the
designer envisaged. As such, the practicality of the site operations should also be
considered, thereby minimising the risk of construction errors.)
Over the years, the calculation of drift loading and associated purlin design
has been made relatively straightforward by the major purlin manufacturers, a
majority of whom offer state of the art software to facilitate rapid design, invariably free of charge.
1.3.2 Wind loads
A further significant change that must be accounted for in the design (in the UK)
of structures in general (including the single-storey structures to which this chapter
alludes) has been the inception of BS 6399: Part 2 – Code of practice for wind loads
in lieu of CP3: Chapter V: Part 2, which has been declared obsolescent.
A cursory inspection of the former will show that BS6399: Part 2 addresses the
calculation of the wind loading in a far more rigorous way than CP3: Chapter V:
Part 2, and offers two alternative methods for determining the loads that the
structure must withstand:
•
•
Standard method – this method uses a simplified procedure to obtain a standard
effective wind speed, which is used with standard pressure coefficients to determine the wind loads for orthogonal design cases
Directional method – this method derives wind speeds and pressure coefficients
for each wind direction, either orthogonal or oblique.
In both methods, the dynamic wind pressure, qs, is calculated as follows:
qs = 0.613 Ve2
Ve = Vs ¥ Sb
Steel Designers' Manual - 6th Edition (2003)
10
Single-storey buildings
Vs = Vb ¥ Sa ¥ Sd ¥ Ss ¥ Sp
Vb = the basic wind speed
Sa = an altitude factor
Sd = a direction factor
Ss = a seasonal factor
Sp = a probability factor
Vs = the site wind speed
Ve = the effective wind speed
Sb = a terrain factor
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
The internal and external pressures that are applied to the structure are calculated
from the generic equation:
p = qs ¥ Cp ¥ Ca
p = either the internal or external applied pressure
Cp = either the internal or external pressure coefficient
Ca = the size effect factor for either internal or external pressures
BS 6399: Part 2 recognizes both site topography and location, in either town or
country, the latter being influenced by the distance to the sea.
The reader will note that the size effect factor was not present in CP3: Chapter
V: Part 2, and the calculation of this factor alone is worthy of further mention.
The size effect factor, Ca, is dependent upon a ‘diagonal dimension – a’, which
varies for each loaded element. BS 6399: Part 2 recognizes the fact that elements
with large diagonal dimensions can have the load to which they may be subject
‘reduced’ from the sum of the design loading of each of the elements that they
support.
For example, the ‘diagonal dimension – a’ for the rafter of a typical portal frame
shown in Fig. 1.7 below is greater than that for each purlin it supports. (This is consistent with the method of BS 6399: Part 1, which allows a percentage reduction in
imposed load on a floor beam, say, depending on the tributary loading area of the
beam.)
However, many purlin manufacturers have updated their design software to
incorporate the requirements of this code of practice, and most of the somewhat
complicated analysis is automatically executed.
BS 6399: Part 2 contains an abundance of tables and graphs from which external
pressure coefficients for many types of building can be determined, and recognises
the fact that certain areas of the structure (adjacent to the eaves, apex and corners,
for example) must be designed to allow for high, local pressure coefficients. This
fact in itself implies that the number of secondary elements such as purlins and
side-rails may increase over those that were required under the criteria of CP3:
Chapter V: Part 2.
As mentioned above, BS 6399: Part 2 offers a rigorous approach to the calculation of wind loads to structures, and a more detailed treatment of this topic is outside
the scope of this chapter. It is hoped that the somewhat brief treatment above will
induce the reader to study BS 6399: Part 2 in some depth, and perhaps calibrate any
Steel Designers' Manual - 6th Edition (2003)
Loading
11
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
a - rafter
Fig. 1.7
Diagonal dimension ‘a’ for purlins and rafters
designs undertaken by hand calculation with those obtained from the manufacturers’ software, to get a ‘feel’ for this new code of practice.
1.3.3 Internal gravity loads
Service loads for lighting, etc., are reasonably included in the global 0.6 kN/m2. As
service requirements have increased, it has become necessary to consider carefully
the provision to be made.
Most purlin manufacturers can provide proprietary clips for hanging limited point
loads to give flexibility of layout. Where services and sprinklers are required, it is
normal to design the purlins for a global service load of 0.1–0.2 kN/m2 with a reduced
value for the main frames to take account of likely spread. Particular items of plant
must be treated individually. The specifying engineer should make a realistic assessment of the need as the elements are sensitive, and while the loads may seem low,
they represent a significant percentage of the total and affect design economy
accordingly.
1.3.4 Cranes
The most common form of craneage is the overhead type running on beams
supported by the columns. The beams are carried on cantilever brackets or in
heavier cases by providing dual columns.
Steel Designers' Manual - 6th Edition (2003)
12
Single-storey buildings
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
In addition to the weight of the cranes and their load, the effects of acceleration
and deceleration have to be catered for. This is simplified by a quasi-static approach
with enhanced load factors being used. The allowances to be made are given in
BS 6399: Part 1. For simple forms of crane gantry these are:
(1) for loads acting vertically, the maximum static wheel loads increased by 25%
for an electric overhead crane or 10% for a hand-operated crane;
(2) for the horizontal force acting transverse to the rails, the following percentage
of the combined weight of the crab and the load lifted:
(a) 10% for an electric overhead crane, or
(b) 5% for a hand-operated crane;
(3) for the horizontal forces acting along the rails, 5% of the static wheel loads
which can occur on the rails for overhead cranes which are either electric or
hand-operated.
For heavy, high-speed or multiple cranes the allowances should be specially
calculated with reference to the manufacturers.
The combination load factors for design are given in BS 5950: Part 1. The constant movement of a crane gives rise to a fatigue condition. This is, however,
restricted to the local areas of support, i.e. the crane beam itself, the support bracket
and its connection to the main column. It is not normal to design the whole frame
for fatigue as the stress levels due to the crane travel are relatively low.
1.3.5 Notional horizontal forces
During the construction of any structure, there will always be the possibility that
practical imperfections such as lack of verticality of the columns, for example, will
be evident. To allow for this eventuality, BS 5950: Part 1 dictates that when considering the strength of the structure, all structures should be capable of resisting
notional horizontal forces (NHFs) equivalent to 0.50% of the factored dead and
imposed loads applied at the same level.
These NHFs should be assumed to act in any one direction at a time, and should
be applied at each roof and floor level or their equivalent. They should be taken as
acting simultaneously with the factored vertical dead and imposed loads, but need
not be applied to a structure that contains crane loading, as the loads from the latter
already contain significant horizontal loads.
The NHFs should not:
•
•
•
•
•
Be applied when considering overturning
Be applied when considering pattern loads
Be combined with applied horizontal loads
Be combined with temperature effects
Be taken to contribute to the net reactions at the foundations.
Steel Designers' Manual - 6th Edition (2003)
Design of common structural forms
13
In up-to-date analysis packages, the application of these NHFs is invariably allowed
for when the load cases and load combinations are being derived.
In addition to the use of NHFs in the assessment of frame strength, they are also
used in ascertaining the stiffness of frames relevant to in-plane stability checks. This
latter consideration is covered in greater detail in 1.4.1, below.
1.4 Design of common structural forms
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
1.4.1 In-plane stability
Without doubt, one of the most significant changes in the recent amendment to
BS 5950: Part 1 is the provision of in-plane stability checks to both multi-storey and
moment-resisting portal frames.
The somewhat simple procedures of Section 2 of BS 5950: Part 1 cannot be
utilized when considering the in-plane stability of portal frames since they do not
consider axial compression within the rafters. Axial compression in the rafters has
a significant effect on the stability of the frame as a whole.
In-plane stability checks are required to ensure that the load that would cause
buckling of the frame as a whole is greater than the sum of the applied forces.
Unlike a beam and column structure, a single-storey, moment-resisting portal
frame does not generally have any bracing in the plane of the frame. As such,
restraint afforded to individual columns and rafters is a function of the stiffness of
the members to which they connect. In simplistic terms, rafters rely on columns,
which in turn rely on rafters. The stability check for the frame must therefore
account for the stiffness of the frame itself.
When any structure is loaded it deflects. To this end, the deflected shape is different from the idealized representation of the frame when it is deemed unloaded
or ‘at rest’. If a frame is relatively stiff, such deflections are minimized. If, however,
a frame is of such a ‘small’ stiffness as to induce significant deflections when loaded,
‘second-order’ effects impact on both the frame’s stiffness and the individual
members’ ability to withstand the applied load.
Consider the example of a horizontal, axially loaded strut as shown in Fig. 1.8.
Prior to application of the axial force, the strut would deflect under its self-weight.
If the strut was relatively stiff, the self-weight deflection, D, would be small. On application of an axial force, P, a bending moment equal to P.D would be induced at
I
P
Fig. 1.8
Horizontal, axially loaded strut
P
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
14
Single-storey buildings
mid-span. The strut would need to be designed for the combined effects of axial
force and bending moment. In the case in question, axial load would have the
greater influence on the design.
If, however, the strut were of a stiffness such that the initial self-weight deflection
was relatively large, the induced second-order effect, P.D, would significantly
increase, and it is possible that the bending moment would play a greater part in
the design of the member, since a greater deflection would induce a larger secondorder bending moment, which in turn would induce a further deflection and a
further second-order moment, etc.
In relation to the deflections of portal frames under load, the above phenomenon is more than sufficient to reduce the frame’s ability to withstand the applied
loads.
BS 5950: Part 1 addresses the problem of in-plane stability of portal frames by
the use of one of the following methods:
(1) the sway check method with the snap-through check
(2) the amplified moments method
(3) second-order analysis.
1.4.1.1 The sway check method
This particular check is the simplest of those referred to above, in so far as elastic
analysis can be used to determine the deflections at the top of the columns due to
the application of the NHFs previously mentioned.
If, under the NHFs of the gravity load combination, the horizontal deflections at
the tops of the columns are less than height/1000, the load factor for frame stability to be satisfied, lr, can be taken as 1.0.
The code also offers the possibility that for frames not subject to loads from either
cranes, valley beams or other concentrated loads, the height/1000 criterion can be
satisfied by reference to the Lb/D ratio, where Lb is the effective span of the frame,
and D the cross-sectional depth of the rafter.
It should be noted, however, that this particular check can only be applied to
frames that fall within certain geometric limits.
The reader is referred to both BS 5950: Part 1 Section 5.5.4.2 and reference [5]
for a more in-depth treatment of the above.
1.4.1.2 The amplified moments method
Where the frame does not meet the criteria of the sway check method, the
amplified moments method referred to in BS 5950: Part 1 Section 5.5.4.4 may be
used.
Steel Designers' Manual - 6th Edition (2003)
Design of common structural forms
15
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
The basis of this method revolves around the calculation of a parameter known
as the lowest elastic critical load factor, lcr, for a particular load combination. (N.B.
BS 5950: Part 1 does not give a method for calculating lcr, but refers to reference
[5].)
A detailed treatment of the calculation of lcr is outside the scope of this particular chapter, and the reader is again referred to reference [5]. Fortunately, the industry software to carry out the analysis of portal frames is capable of calculating lcr
to BS 5950: Part 1. Accordingly, the task is not as daunting as first appears. However,
it is imperative that the reader understands the background to this particular
methodology.
On determining lcr, the required load factor, lr, is calculated as follows:
lcr 10 : lr = 1.0
10 > lcr 4.6 : lr = 0.9lcr/(lcr - 1)
If lcr < 4.6, the amplified moments method cannot be used and a second-order
analysis should be carried out.
The application of lr in the design process is as follows:
•
•
For plastic design, ensure that the plastic collapse factor, lp lr, and check the
member capacities at this value of lr.
For elastic design, if lr > 1.0, multiply the ultimate limit state moments and forces
arrived at by elastic analysis by lr, and check the member capacities for these
‘amplified’ forces.
1.4.1.3 Second-order analysis
Second-order analysis, briefly referred to above, accounts for additional forces
induced in the frame due to the axial forces acting eccentrically to the assumed
member centroids as the frame deflects under load.
These secondary effects, often referred to as ‘P-Delta’ effects, can be best illustrated by reference to Fig. 1.9 of a simple cantilever.
As can be seen, the second-order effects comprise an additional moment of PD
due to the movement of the top of the cantilever, D, induced by the horizontal force,
H, in addition to a moment within the member of Pd due to deflection of the
member itself between its end points. (It should be noted that, in certain instances,
second-order effects can be beneficial. Should the force P, above, have been tensile,
the bending moment at the base would have been reduced by PD kNm.)
In the case of a portal frame, there are joint deflections at each eaves and apex,
together with member deflections in each column and rafter. In the case of the
columns and rafters, it is standard practice to divide these members into several subelements between their start and end nodes to arrive at an accurate representation
of the secondary effects due to member deflections.
As a consequence of these effects, the stiffness of the portal frame is reduced
Steel Designers' Manual - 6th Edition (2003)
16
Single-storey buildings
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
P
Fig. 1.9
Second-order effects in a vertical cantilever
below that arrived at from a first-order elastic analysis. Again, the industry software
packages include modules for the rapid calculation of these second-order effects,
and utilize either the matrix or energy methods of analysis to arrive at the required
solution, in most cases employing an iterative solution.
1.4.1.4 Tied portals
The reader’s attention is drawn to the fact that for tied portals, it is mandatory to
use a second-order analysis to check for in-plane stability. In the case of the tied
portal under gravity loading, the tie is subject to quite high tensile loads and, as
such, would elongate under load. This in turn would cause the eaves to spread, with
a subsequent downward deflection of the apex. This downward deflection of the
apex reduces the lever arm between the rafter and the tie, and as such increases the
tensile force in the tie and the compressive force in the rafter. A first-order elastic
analysis would not truly represent what actually occurs in practice, and, subject
to the dimensions of the tie, may be somewhat unsafe in arriving at the design
forces to which the frame is subjected.
1.4.2 Beam and column
The cross-section shown in Fig. 1.4 is undoubtedly the simplest framing solution
which can be used to provide structural integrity to single-storey buildings. Used
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Design of common structural forms
17
predominantly in spans of up to 10 m, where flat roof construction is acceptable, the
frame comprises standard hot-rolled sections having simple or moment-resisting
joints.
Flat roofs are notoriously difficult to weatherproof, since deflections of the horizontal cross-beam induce ponding of rainwater on the roof, which tends to penetrate the laps of traditional cladding profiles and, indeed, any weakness of the
exterior roofing fabric. To counteract this, either the cross-member is cambered to
provide the required fall across the roof, or the cladding itself is laid to a predetermined fall, again facilitating drainage of surface water off the roof.
Due to the need to control excessive deflections, the sections tend to be somewhat heavier than those required for strength purposes alone, particularly if the
cross-beam is designed as simply-supported. In its simplest form, the cross-beam is
designed as spanning between columns, which, for gravity loadings, are in direct
compression apart from a small bending moment at the top of the column due to
the eccentricity of the beam connection. The cross-beam acts in bending due to the
applied gravity loads, the compression flange being restrained either by purlins,
which support the roof sheet, or by a proprietary roof deck, which may span between
the main frames and which must be adequately fastened. The columns are treated
as vertical cantilevers for in-plane wind loads.
Resistance to lateral loads is achieved by the use of a longitudinal wind girder,
usually situated within the depth of the cross-beam. This transmits load from the
top of the columns to bracing in the vertical plane, and thence to the foundation.
The bracing is generally designed as a pin-jointed frame, in keeping with the simple
joints used in the main frame. Details are shown in Fig. 1.10.
Buildings which employ the use of beam-and-column construction often have
brickwork cladding in the vertical plane. With careful detailing, the brickwork can
be designed to provide the vertical sway bracing, acting in a similar manner to the
shear walls of a multi-storey building.
gable end
vertical bracing
Fig. 1.10
Simple wind bracing system
Steel Designers' Manual - 6th Edition (2003)
18
Single-storey buildings
Resistance to lateral loading can also be achieved either by the use of rigid
connections at the column/beam joint or by designing the columns as fixed-base
cantilevers. The latter point is covered in more detail in the following subsection relating to the truss and stanchion framing system.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
1.4.3 Truss and stanchion
The truss and stanchion system is essentially an extension of the beam-and-column
solution, providing an economic means of increasing the useful span.
Typical truss shapes are shown in Fig. 1.11.
Members of lightly-loaded trusses are generally hot-rolled angles as the web
elements, and either angles or structural tees as the boom and rafter members,
the latter facilitating ease of connection without the use of gusset plates. More
heavily loaded trusses comprise universal beam and column sections and hot-rolled
channels, with connections invariably employing the use of heavy gusset plates.
Fig. 1.11
Truss configurations
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Design of common structural forms
19
In some instances there may be a requirement for alternate columns to be omitted
for planning requirements. In this instance load transmission to the foundations is
effected by the use of long-span eaves beams carrying the gravity loads of the intermediate truss to the columns: lateral loading from the intermediate truss is transmitted to points of vertical bracing, or indeed vertical cantilevers by means of
longitudinal bracing as detailed in Fig. 1.12. The adjacent frames must be designed
for the additional loads.
Considering the truss and stanchion frame shown in Fig. 1.13, the initial assumption is that all joints are pinned, i.e. they have no capacity to resist bending moment.
The frame is modelled in a structural analysis package or by hand calculation, and,
for the load cases considered, applied loads are assumed to act at the node points.
It is clear from Fig. 1.13 that the purlin positions and nodes are not coincident;
consequently, due account must be taken of the bending moment induced in the
rafter section. The rafter section is analysed as a continuous member from eaves to
apex, the node points being assumed as the supports, and the purlin positions as the
points of load application (Fig. 1.14).
The rafter is sized by accounting for bending moment and axial loads, the web
members and bottom chord of the truss being initially sized on the basis of axial
load alone.
Use of structural analysis packages allows the engineer to rapidly analyse any
number of load combinations. Typically, dead load, live load and wind load cases are
analysed separately, and their factored combinations are then investigated to determine the worst loading case for each individual member. Most software packages
provide an envelope of forces on the truss for all load combinations, giving
truss for horizontal load
T/VNJNV
truss for vertical load
Fig. 1.12
Additional framing where edge column is omitted
Steel Designers' Manual - 6th Edition (2003)
20
Single-storey buildings
B
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 1.13
Truss with purlins offset from nodes
A1
node points
E load application points
Fig. 1.14
Rafter analysed for secondary bending
maximum tensile and compressive forces in each individual member, thus facilitating rapid member design.
Under gravity loading the bottom chord of the truss will be in tension and the
rafter chords in compression. In order to reduce the slenderness of the compression
members, lateral restraint must be provided along their length, which in the present
case is provided by the purlins which support the roof cladding. In the case of load
reversal, the bottom chord is subject to compression and must be restrained. A
typical example of restraint to the bottom chord is the use of ties, which run the
length of the building at a spacing governed by the slenderness limits of the compression member; they are restrained by a suitable end bracing system. Another
solution is to provide a compression strut from the chord member to the roof
purlin, in a similar manner to that used to restrain compression flanges of rolled
sections used in portal frames. The sizing of all restraints is directly related to the
compressive force in the primary member, usually expressed as a percentage of
the compressive force in the chord. Care must be taken in this instance to ensure
that, should the strut be attached to a thin-gauge purlin, bearing problems in thingauge material are accounted for. Examples of restraints are shown in Fig. 1.15.
Connections are initially assumed as pins, thereby implying that the centroidal
axes of all members intersecting at a node point are coincident. Practical considerations invariably dictate otherwise, and it is quite common for member axes to be
eccentric to the assumed node for reasons of fit-up and the physical constraints that
Steel Designers' Manual - 6th Edition (2003)
Design of common structural forms
21
purl in
,
7/
main
truss
restraint
flonge
fi an ge
VmPresson
longitudinal
angle
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 1.15
Restraints to bottom chord members
are inherent in the truss structure. Such eccentricities induce secondary bending
stresses of the node points, which must be accounted for not only by local bending
and axial load checks at the ends of all constituent members, but also in connection
design. Typical truss joints are detailed in Fig. 1.16.
It is customary to calculate the net bending moment at each node point due to
any eccentricities, and proportion this moment to each member connected to the
node in relation to member stiffness.
In heavily-loaded members secondary effects may be of such magnitude as to
require member sizes to be increased quite markedly above those required when
considering axial load effects alone. In such instances, consideration should be given
to the use of gusset plates, which can be used to ensure that member centroids are
coincident at node points, as shown in Fig. 1.17. Types of truss connections are very
much dependent on member size and loadings. For lightly-loaded members, welding
is most commonly used with bolted connections in the chords if the truss is to be
transported to site in pieces and then erected. In heavily-loaded members, using
gusset plates, either bolting, welding or a combination of the two may be used.
However, the type of connection is generally based on the fabricator’s own
reference.
Where the roof truss has a small depth at the eaves, lateral loading is resisted
either by longitudinal wind girders in the plane of the bottom boom and/or rafter,
or by designing the columns as vertical fixed-base cantilevers, as shown in Fig. 1.18.
Where the truss has a finite depth at the eaves, benefit can be obtained by developing a moment connection at this position with the booms designed for appropriate additional axial loads. This latter detail may allow the column base to be
designed as a pin, rather than fixed, depending on the magnitude of the applied
loading, and the serviceability requirements for deflection.
Longitudinal stability is provided by a wind girder in the plane of the truss boom
and/or rafter at the gable wall, the load from the gable being transmitted to the
foundations by vertical bracing as shown in Fig. 1.19.
Typical joints in trusses
22
Fig. 1.16
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
Single-storey buildings
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Design of common structural forms
Fig. 1.17
Ideal joint with all member centroids coincident
Fig. 1.18
Sway resistance for truss roofs
____
plan
Fig. 1.19
Gable-end bracing systems
/
elevation
23
Steel Designers' Manual - 6th Edition (2003)
24
Single-storey buildings
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
1.4.4 Portal frames
By far the most common form of structure for single-storey structures is the portal
frame, the principal types being shown in Fig. 1.20.
Spans of up to 60 m can be achieved by this form of construction, the frame generally comprising hot-rolled universal beam sections. However, with the increase
in understanding of how slender plate elements react under combined bending
moment, axial load and shear force, several fabricators now offer a structural frame
fabricated from plate elements. These frames use tapered stanchions and rafters to
provide an economic structural solution for single-storey buildings, the frame being
‘custom designed’ for each particular loading criterion.
Roof slopes for portal frames are generally of the order of 6° but slopes as low
as 1° are becoming increasingly popular with the advent of new cladding systems
such as standing seam roofs. It should be noted, however, that frame deflections at
low slopes must be carefully controlled, and due recognition must be taken of the
large horizontal thrusts that arise at the base.
Frame centres are commonly of the order of 6–7.5 m, with eaves heights ranging
from 6 to 15 m in the case of aircraft hangars or similar structures.
Resistance to lateral loading is provided by moment-resisting connections at the
eaves, stanchion bases being either pinned or fixed. Frames which are designed on
the basis of having pinned bases are heavier than those having fixity at the bases,
although the increase in frame cost is offset by the reduced foundation size for the
pinned-base frames.
II
(a)
(c)
Fig. 1.20
Portal-frame structures
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Design of common structural forms
25
Parallel-flange universal sections, subject to meeting certain physical constraints
regarding breadth-to-thickness ratios of both flanges and webs, lend themselves to
rapid investigation by the plastic methods of structural analysis. The basis of the
plastic method is the need to determine the load applied to the frame which will
induce a number of ‘plastic hinges’ within the frame, thereby causing failure of the
frame as a mechanism.
This requirement is best illustrated by the following simple example.
Considering the pinned-base frame shown in Fig. 1.21(a), subject to a uniform
vertical load, w, per unit length: the reactions at the foundations are shown in
Fig. 1.21(b). The frame has one degree of indeterminacy. In order that the frame
fails as a mechanism, at least two plastic hinges must form (i.e. the degree of
indeterminacy + 1) as shown in Fig. 1.21(c). (It should be noted that although four
hinges are shown in Fig. 1.21(c), due to ‘theoretical’ symmetry only one pair either
side of the apex will in fact form, due to the obvious imperfections in both loading
and erection conditions.)
In many structures, other than the most simple, it is not clear where the plastic
hinges will form. There are several methods available to the design engineer which
greatly assist the location of these hinge positions, not least the abundance of proprietary software packages specifically relating to this form of analysis. Prior to the
use of these packages, however, it is imperative that the engineer fully understands
the fundamentals of plastic analysis by taking time to calculate, by hand, several
design examples. The example which follows uses a graphical construction as a
means of illustrating the applications of the method to a simple portal frame.
Further information on plastic analysis is given in Chapter 11.
The frame shown in Fig. 1.22(a) has one degree of indeterminacy. It is made
statically determinate by assuming a roller at the right-hand base as shown in
Fig. 1.22(b), and the free-bending moment diagram drawn as shown in Fig. 1.22(c).
The reactant line for the horizontal force ‘removed’ to achieve a statically determinate structure must now be drawn as follows:
w/unit length
plastic hinges
V
(a)
Fig. 1.21
V
(b)
Structural behaviour of pinned-base portal
(c)
Steel Designers' Manual - 6th Edition (2003)
Single-storey buildings
26
w/unit length
I
I
C
:fl:
_______
h,
________
L1
V
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(a)
V
(b)
hinge positions
E
0
B
(c)
Fig. 1.22
Application of graphical method
(1) Considering member DC, the bending moment due to the reaction R at point
D, MD = Rh1 . Similarly, at point C, MC = R(h1 + h2 ).
(2) The gradient of the reactant line along member DC:
m=
R(h1 + h2 ) - R(h1 ) Rh2
=
L1
L1
(3) The reactant line must past through a point O as shown in Fig. 1.22(c). By similar
triangles:
Rh1 Rh2
=
x
L1
giving
x=
L1 h1
h2
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Design of common structural forms
27
Therefore, by rotating the reactant line through point O, the positions of the bending
moment of equal magnitude in the positive and negative regions can be found and
the member sized accordingly based on this value of bending moment.
Having found the positions of the reactant line which gives the number of hinges
required for a mechanism to form, in this case two, the reactant line for the stanchions can be drawn through point O, a distance h1 from the end of the free-bending
moment diagram. The unknown reaction, H, is then calculated as H = MD/h1.
In a majority of instances, portal frames are constructed with a haunch at the
eaves, as shown in Fig. 1.23(a).
Depending on the length/depth of the haunch, the plastic hinges required for a
mechanism to form are shown in Figs. 1.23(b) and (c). The dimensional details for
the haunch can be readily investigated by the graphical method by superimposing
the dimensions of the haunch on the free-bending moment diagram. The reactant
line can then be rotated accordingly until the required mechanism is achieved and
members sized accordingly.
Haunches are generally fabricated from parallel beam sections (Fig. 1.24). In all
cases, the haunch must remain in the elastic region. A detailed check is required
along its length, from a stability point of view, in accordance with the requirements
laid down in the relevant code of practice. In some instances, bracing of the bottom flange must be provided from a purlin position within its length, as shown in
Fig. 1.25.
haunch
(a)
Fig. 1.23
Fig. 1.24
(c)
(b)
Alternative hinge locations for haunched portal frames
V /H-
Haunch fabrication
Cut tines
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
28
Fig. 1.25
Single-storey buildings
Bottom flange restraint
Portal frames analysed by the plastic methods of structural analysis tend to be
more economical in weight than their elastically designed counterparts. However,
engineers should be aware that minimum weight sections, and by inference
minimum depth sections, have to be connected together to withstand the moments
and forces induced by the applied loading. Particular attention should be paid at an
early stage in the design process to the economics of the connection. Cost penalties
may be induced by having to provide, for example, gusset plates between pairs of
bolts should the section be so shallow as to necessitate a small number of highlystressed bolts. In addition, end plates should not be much thicker than the flanges
of the sections to which they are attached.
Provided the engineer is prepared to consider the implications of his calculated
member sizes on the connections inherent in the structure at an early stage, an
economic solution will undoubtedly result. Leaving connection design solely to the
fabricator, without any consideration as to the physical constraints of providing
Steel Designers' Manual - 6th Edition (2003)
Design of common structural forms
29
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
a number of bolts in an extremely shallow depth, will undoubtedly result in a connection which is both difficult to design and fabricate, and costly.
Having sized the members based on the previous procedure, it is imperative that
an analysis at serviceability limit state is carried out (i.e. unfactored loads) to check
deflections at both eaves and apex. This check is required not only to ascertain
whether deflections are excessive, but also as a check to ensure that the deflections
and accompanying frame movement can be accommodated by the building envelope without undue cracking of any brickwork or tearing of metal cladding sheets
at fixing positions. Excessive lateral deflections can be reduced by increasing the
rafter size and/or by fixing the frame bases. It should be noted that the haunch has
a significant effect on frame stiffness due to its large section properties in regions
of high bending moment.
1.4.5 Tied portal
The tied portal, a variation of the portal frame, is illustrated in Fig. 1.26.
Economy of material can be achieved, albeit at the expense of reducing the allowable headroom of the building, by provision of a tie at eaves level. Under vertical
loading, the eaves spread is reduced due to the induced tensile force in the tie.
However, although the rafter size is reduced, because under vertical loading the only
possible mode of failure of the rafter is that of a fixed-end beam, deflections are
more critical. Lateral loading due to wind further complicates the problem since in
some cases the tie may not act in tension and becomes redundant. It is common,
therefore, for tied portals to have fixed bases, which provide greater stability and
resistance to horizontal loading.
1.4.6 Stressed-skin design
In addition to providing the weathertight membrane, the steel sheeting can also be
utilized as a structural element itself. Correct detailing of connections between individual cladding sheets (and their connections to the support steelwork) will induce
a stressed-skin effect which can offer both resistance to transverse wind loads, and
restraint to the compression flanges of the main frame elements.
Fig. 1.26
Tied portal frame
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
30
Single-storey buildings
It is clear that, when vertical load is applied to a pitched roof portal frame,
the apex tends to deflect downwards, and the eaves spread horizontally. This displacement cannot occur without some deformation of the cladding. Since the
cladding is fixed to the purlins, the behaviour of the cladding and purlins together
is analogous to that of a deep plate girder, i.e. the purlins resist the bending moment,
in the form of axial forces, and the steel sheeting resists the applied shear, as shown
in Fig. 1.27.
As a consequence of this action, the load applied over the ‘length’ of the ‘plate
girder’ has to be resisted at the ends of the span. In the case of stressed-skin action
being used to resist transverse wind loads on the gable end of the structure, adequate connection must be present over the end bay to transmit the load from the
assumed ‘plate girder’ into the braced bay, and into the foundations, as shown in
Fig. 1.28.
Some degree of stressed-skin action is present in all portal frame structures where
cladding is fixed to the supporting members by mechanical fasteners. Claddings
which are either brittle (i.e. fibre cement) or are attached to the supporting
structure by clips (i.e. standing seam systems) are not suitable for stressed-skin
applications.
sheeting (web)
purlins (flange)
Fig. 1.27
Stressed-skin action
shear panel in sheeting
adequate
connections
required here
Fig. 1.28
Stressed-skin action for gable-end bracing
Steel Designers' Manual - 6th Edition (2003)
Design of common structural forms
31
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Correct detailing and correct fixing is essential to the integrity of stressed-skin
applications:
(1) Suitable bracing members must be used to allow the applied loads from
diaphragm action to be transmitted to the foundations.
(2) At the position of laps in the sheeting, suitable mechanical fasteners must be
present to allow continuity of load between sheets. These fasteners can be
screws or rivets, which must be capable of withstanding both shear and pull out
due to the stressed-skin effect and applied loads respectively.
(3) The ends of the sheet must similarly be connected to the supporting members
(i.e. purlins).
(4) All stressed-skin panels must be connected to adequately designed edge
members, which must be capable of transmitting the axial forces induced by
bending, in addition to the forces induced by imposed and wind load effects.
1.4.7 Purlins and siderails
Due to the increasing awareness of the load-carrying capabilities of sections formed
from thin-gauge material, proprietary systems for both purlins and siderails have
been developed by several manufacturers in the United Kingdom. Consequently,
unless a purlin or siderail is to be used in a long-span or high-load application (when
a hot-rolled angle or channel may be used), a cold-formed section is the most frequently used cladding support member for single-storey structures.
Cold-formed sections manufactured from thin-gauge material are particularly
prone to twisting and buckling due to several factors which are directly related to
the section’s shape. The torsional constant of all thin-gauge sections is low (it is a
function of the cube of the thickness); in the case of lipped channels the shear centre
is eccentric to the point of application of load, thus inducing a twist on the section;
in the case of Zeds the principal axes are inclined to the plane of the web, thus
inducing bi-axial bending effects. These effects affect the load-carrying capacity of
the section.
When used in service the support system is subject to downward loading due to
dead and live loads such as cladding weight, snow, services, etc., and uplift if the
design wind pressure is greater than the dead load of the system. Therefore, for
a typical double-span system as shown in Fig. 1.29, the compression flange is
restrained against rotation by the cladding for downward loading, but it is not so
restrained in the case of load reversal.
In supporting the external fabric of the building, the purlins and siderails gain
some degree of restraint against twisting and rotation from the type of cladding used
and the method of its fastening to the supporting member. In addition, the connection of the support member to the main frame also has a significant effect on the
load-carrying capacity of the section. Economical design, therefore, must take
account of the above effects.
Steel Designers' Manual - 6th Edition (2003)
Single-storey buildings
32
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 1.29
WI
Typical double-span purlin with cladding restraint
x
L
Fig. 1.30
collapse load/span
x
:
L
Collapse mechanism for a two-span purlin system
There are four possible approaches to the design of a purlin system:
(1) Design by calculation based on an elastic analysis as detailed in the relevant
code of practice BS 5950: Part 5.6 This approach neglects any beneficial effect
of cladding restraint for the wind uplift case.
(2) Empirical design based on approximate procedures for Zeds as given in the
codes of practice. This approach leads to somewhat uneconomic design.
(3) Design by calculation based on a rational analysis which accounts for the
stabilizing influence of the cladding, plasticity in the purlin as the ultimate
load is approached, and the behaviour of the cleat at the internal support. The
effects, however, are difficult to quantity.
(4) Design on the basis of full-scale testing.
Manufacturers differ in the methods: the example used here is from one manufacturer who has published the method used.7
For volume production, design by testing is the approach which is used. Although
this approach is expensive, maximum economy of material can be achieved and the
cost of the testing can be spread over several years of production.
Design by testing involves the ‘fine-tuning’ of theoretical expressions for the collapse load of the system. The method is based on the mechanism shown in Fig. 1.30
for a two-span system.
From the above mechanism it can be shown that:
Steel Designers' Manual - 6th Edition (2003)
Design of common structural forms
33
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
the collapse load, Wc = f (M1 , M 2 , x, L)
x = f (M1 , M 2 , L)
q 1 = f (Wc , M1 , L)
The performance of, for example, a two-span system is considerably enhanced if
some redistribution of bending moment from the internal support is taken into
account. The moment–rotation characteristic at the support is very much dependent on the cleat detail and the section shape. The characteristics of the central
support can be found by testing a simply-supported beam subject to a central
point load, so as to simulate the behaviour of the central support of a double-span
system.
From this test, the load–deflection characteristics can be plotted well beyond the
deflection at which first yield occurs. A lower bound empirical expression can then
be found for the support moment, M1, based on an upper limit rotational capacity.
A similar expression can be found for M2, the internal span moment, again on the
basis of a test on a simply-supported beam subject to a uniformly distributed load,
applied by the use of a vacuum rig, or perhaps sandbags.
The design expressions can then be confirmed by the execution of numerous fullscale tests on double-span systems employing pairs of purlins supporting proprietary cladding.
As described earlier, load reversal under wind loading invariably occurs, thereby
inducing compressive forces to the flange in the internal span, which is not
restrained by the cladding as is the case in the downward load case. Anti-sag rods
or the like are placed within the internal span, thereby reducing the overall buckling length of the member. The system is again tested in load reversal conditions
and, as before, the design expressions can be further refined.
In some instances, sheeting other than conventional trapezoidal cladding (which
is invariably through fixed to the purlin by self-drilling fastenings in alternate
troughs) will not afford the full restraint to the compression flange: examples are
standing seam roofs, brittle cladding, etc. The amount of restraint afforded by these
latter types of cladding cannot easily be quantified. For the reasons outlined above,
further full-scale testing and similar procedures of verification of design expressions
are carried out.
The results of the full-scale tests are then condensed into easy to use load–span
tables which are given in the purlin manufacturers’ design and detail literature. The
use of these tables is outlined in Table 1.1.
The tabular format is typical of that contained in all purlin manufacturers’ technical literature. The table is generally prefaced by explanatory notes regarding fixing
condition and lateral restraint requirements, the latter being particularly relevant
to the load-reversal case. Conditions which arise in practice, and which are not
covered in the technical literature, are best dealt with by the manufacturers’ technical services department, which should be consulted for all non-standard cases.
Siderail design is essentially identical to that for purlins: load capacities are again
arrived at after test procedures.
Self-weight deflections of siderails due to bending about the weak axis of the
section are overcome by a tensioned wire system incorporating tube struts, typically
Steel Designers' Manual - 6th Edition (2003)
34
Single-storey buildings
Table 1.1 Typical load table
Span (m)
6.0
Section
CF170160
CF170170
CF170180
CF200160
CF200180
UDL (kN)a
9.50
10.75
11.75
12.25
13.50
Purlin centres (mm)
1000
1200
1400
1600
1800
1.58
1.79
1.96
2.04
2.25
1.32
1.49
1.63
1.70
1.88
1.13
1.28
1.40
1.46
1.61
0.99
1.12
1.22
1.28
1.41
0.88
1.00
1.09
1.13
1.25
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
a
UDL is the total uniformly distributed load on a single purlin of a 6 m double-span system which would
produce a failure of the section.
Loads given beneath the columns under purlin centres give the allowable uniformly distributed load on
the purlin at a given spacing. The figure is arrived at by dividing the UDL by the product of span and purlin
centres, i.e. for CF170180, 6 m span, 1800 mm centre,
allowable load = 11.75/ (6 ¥ 1.8) = 1.09 kN/m2
wire ropes
Fig. 1.31
Anti-sag systems for side wall rails
at mid-span for spans of 6–7 m, and third-points of the span for spans of 7–8 m and
above (Fig. 1.31).
1.4.8 Cladding
In addition to visual and aesthetic requirements, one of the fundamental roles
of the cladding system is to provide not only the weatherproofing for the building but also the insulation requirements suited to both the building and the
environment.
Cladding profiles are roll-formed and produced in volume, so giving excellent
economy, and this means that, apart from an extremely rare occasion, the chosen
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Design of common structural forms
35
profile will be from a manufacturer’s published range. The loads that the profile can
withstand are determined by the manufacturer and published in his brochure. Until
recently they would have been determined by test, but calculation methods have
been developed which are satisfactory for trapezoidal shapes. These give slightly
conservative answers but, in general, profiles that are walkable will have capacities
in excess of those required. Sheeting can be erected satisfactorily without the need
to walk indiscriminately over the whole area. The crowns at mid-span are the most
susceptible to foot traffic but these positions can easily be avoided, even if crawl
boards are not used.
In double-skin construction it is normal to assume all loads are taken on the outer
sheet, and the inner liner merely has to be erectable and stiff enough to prevent
noticeable sag once installed.
The specifier, therefore, has to select the profile appropriate to the design
requirement and check the strength and deflection criteria against the manufacturer’s published load tables. Care must also be taken to ensure the fasteners are
adequate.
As mentioned briefly in 1.2.1 above, the use of higher specification cladding
systems, by default, reduces the emissions of greenhouse gases; the pressing need to
reduce such gases is an almost internationally accepted goal.
In recognition of this, the UK pledged, at the Kyoto summit, to reduce greenhouse gases by some 12% by 2010. In addition, the government’s own manifesto
included a commitment to reduce CO2 emissions by 20%.
Consequently, in order to achieve this goal, significant changes have been proposed to the Building Regulations (England & Wales) – Approved Document Part
L for Non-Domestic Buildings,8 since it was recognized that a significant proportion
of the UK’s emission of greenhouse gases arises from energy used in the day-to-day
use of building structures.
The significant changes are outlined below:
•
•
•
•
•
•
Increase thermal insulation standards by use of improved U values
Introduce an air leakage index
Introduce as-built inspections
Introduce whole building design by integrating the building fabric with heating,
cooling and air-conditioning requirements
Introduce the monitoring of material alterations to an existing structure
Introduce operating log books and energy consumption meters.
1.4.8.1 Improved U values
Approved Document Part L embraces significant changes to the insulation requirements of the building fabric that will impact on the use of fuel and power.
The intended reduction in the U values of both roofs and walls is detailed
below.
Steel Designers' Manual - 6th Edition (2003)
36
Single-storey buildings
Roof systems
Current
Part L
2007
U Value
0.45 W/m2K
0.35 W/m2K
0.16 W/m2K
Current
Part L
2007
0.45 W/m2K
0.35 W/m2K
0.25 W/m2K
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Wall systems
As a consequence of the above, the insulation thickness in either composite, glass
fibre or rock fibre cladding systems will significantly increase. This will have an effect
on the dead load that is applied to the supporting structure, and could result in some
nominal increases in section sizes of the primary and secondary steelwork of the
building.
1.4.8.2 Air leakage index
It will be mandatory for buildings to be designed and constructed so that an air
leakage index will not be exceeded. This index will be set at a specific value of
volume of air leakage per hour per square metre of external surface at a pressure
difference of 50 pascals.
Present proposals for this air leakage index are as follows:
Maximum leakage rate:
Maximum leakage rate:
December 2002
2007
10.0 m3/hr/m2 @ 50 Pa
5.0 m3/hr/m2 @ 50 Pa
To achieve the above, care must be taken to specify and effectively construct sealed
junctions between elements, thereby minimizing air leakage.
The air leakage index will be quantified by in situ air pressurization tests, using
fans, which must always be used on buildings with a floor area that exceeds 1000 m2.
Should the building fail the test, remedial measures must be undertaken, and
further tests carried out until the building is deemed to comply to the satisfaction
of building control.
1.4.8.3 ‘As-built’ inspections
In order that the revised insulation standards are adhered to, it will be necessary to
ensure that the provision of insulation within the fabric itself does not exhibit zones
that could compromise the specification. For example, there should be no signifi-
Steel Designers' Manual - 6th Edition (2003)
Design of common structural forms
37
cant leakage paths between panels; thermal bridges should be minimized; insulation
should be continuous, dry and as ‘uncompressed’ as possible.
One method of ascertaining compliance would be by use of infra-red thermography, where the external fabric of the building is ‘photographed’ using specialist
equipment. By this method, areas of the external fabric that are ‘hot’ – implying
poor insulation and heat loss from the building – will be shown as colours at the
red end of the spectrum. Conversely, areas that are ‘cold’ will be shown at the blue
end of the spectrum.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
1.4.8.4 Whole building design
As the title of this subsection suggests, the building should be designed taking due
account of how the constituent parts interact with each other. To this end, insulation, air tightness, windows, doors and rooflights, heating systems and the like should
not be viewed as individual elements, but rather as elements that, in some way, have
influence on the in-service performance of each other.
1.4.8.5 Material alterations
Material alteration is intended to cover substantial works to existing buildings that
were designed and constructed prior to the changes in Approved Document
Part L.
For example, should major works be required to roof or wall cladding, it will be
necessary to provide insulation to achieve the U value of a new building. Following
on from the latter eventuality will be the need to make provision for improving the
building’s air-tightness.
The replacement of doors and windows will also entail the use of products that
meet the requirements of new buildings, as too will the installation of new heating
systems, for example.
1.4.8.6 Building log books and energy meters
The building owner will initially be provided with details of all the products that
constitute the building, including a forecast of annual energy consumption based
on the building design specification. The owner in turn will be required to keep detailed maintenance records and the like to ensure that the products within the
building are properly maintained, and, when replaced, fully comply with the new
requirements.
Steel Designers' Manual - 6th Edition (2003)
38
Single-storey buildings
Energy meters should be provided to ensure that comparisons of energy consumption can be made with those forecast at design stage. These meters in turn will
provide any new owner/tenant with detailed information on which to base future
energy forecasts.
The cladding has also to withstand the applied loads of snow, wind, and foot traffic
during fixing and maintenance. It must also provide the necessary lateral stability
to the supporting purlin and siderail systems. Occasionally it will form part of the
lateral stability of the structure in the form of a stressed-skin diaphragm, mentioned
above.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
1.4.8.7 Cladding systems
A variety of systems is available to suit environmental and financial constraints. The
most common are listed below.
Single-skin trapezoidal roofing
This was widely used in the past with plasterboard or similar material as the lining
material, and fibreglass insulation in the sandwich. The construction is susceptible
to the plasterboard becoming damp due to condensation. An alternative is the use
of rigid insulation boards, which are impervious to damp, supported on tee bars
between the purlins. Unless the joints are sealed, which is difficult to achieve, condensation is likely to form. Although inexpensive, this type is therefore limited in
its applicability.
The minimum slope is governed by the need to provide watertight joints and
fasteners. If manufacturers’ instructions on the use of sealants and stitching to
laps are rigorously followed, this type can be used down to slopes of approximately 4°.
Double-shell roof construction
In this form of construction the plasterboard has been replaced by a steel liner sheet
of 0.4 mm thickness with some stiffening corrugations. The lining is first installed
and fastened to the purlins, followed by the spacing Zeds, insulation and outer sheet.
The liner tray is not designed to take full wind and erection loads, and therefore
large areas should not be erected in advance of the outer skin. The liner tray is normally supplied in white polyester finish, providing a pleasing internal finish. The
weatherproofing criteria are the same as for single-skin systems and generally the
minimum slope is 4°. Differing thicknesses of insulation are accommodated by
varying spacer depths. The norm is 80 mm of fibreglass giving a nominal U value of
0.44 W/m2 °C.
Steel Designers' Manual - 6th Edition (2003)
Design of common structural forms
39
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Standing seam systems
The traditional forms of construction described above suffer from the inherent disadvantage of having to be fixed by screw-type fasteners penetrating the sheet.
Traditional fixing methods also limit the length of sheet that can be handled even
if, in theory, long lengths can be rolled; thus laps are required.
The need for weathertightness at the lap constrains the minimum slope. A
5000 m2 traditional roof has 20 000 through fasteners and has to resist around
1 million gallons of water a year. The difficulty in ensuring that this large number
of fasteners is watertight demonstrates the desirability of minimizing the number
of penetrations. This has led to the development of systems having concealed fastenings and the ability to roll and fix long lengths. In order to cater for the thermal
expansion in sheets, which may be 30 m long, the fastenings are in the form of clips
which, while holding down the sheeting, allow it to move longitudinally.As discussed
elsewhere, this may reduce the restraint available to the purlins and affect their
design. When used in double-skin configuration the liner panel is normally conventionally fastened and provides sufficient restraint. The available permutations are
too numerous to give general rules but purlin manufacturers will give advice.
It is necessary to fasten the sheets to the structure at one point to resist downslope forces and progressive movement during expansion and contraction. With the
through fasteners reduced to the minimum and laps eliminated or specially detailed,
roof slopes as low as 1° (after deflection) can be utilized. The roofs must be properly maintained since accumulation of debris is more likely and ponding leads to a
reduced coating life.
Standing seam systems are used to replace the traditional trapezoidal outer sheets
in single- and double-skin arrangements as described earlier.
Composite panels
This most recent development in cladding systems provides solutions for many of
the potential problems with metal roofing. The insulating foam is integral with the
sheets and so totally fills the cavity, and with good detailing at the joints condensation can be eliminated in most environments.
The strength of the panel is dependent on the composite action of the two metal
skins in conjunction with the foam. Theoretical calculations are possible although
there are no codified design procedures. Since both steel and foam properties can
vary, and these are predetermined by the manufacturer, it is a question of selecting
the panels from load tables provided rather than individual design. In addition to
having to resist external loads, the effects of temperature differential must be taken
into account. The critical combinations are wind suction with summer temperatures
and snow acting with winter temperatures. The range of temperature considered
is dependent on the colour and hence heat absorption of the outer skin; darker
colours for roofs should only be considered in conjunction with the manufacturer,
if at all.
Steel Designers' Manual - 6th Edition (2003)
40
Single-storey buildings
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Both standing seam and traditional trapezoidal forms are available with the same
slope restrictions as non-composite forms.
A particular advantage is the erectability of the panels, which is a one-pass operation and, therefore, a rapid process. This is combined with inherent robustness and
walkability.
Since the integrity of the panel is important, and it is difficult to inspect the foam
and its adhesion once manufactured, quality control of the materials and manufacturing environment in terms of temperature and dust control is vital. Reputable
manufacturers should, therefore, be specified and their manufacturing methods
ascertained.
External firewall
Where buildings are close to the site boundary the Building Regulations require
that the construction is such that reasonable steps are taken to prevent fire spreading to adjacent property. It has been demonstrated by tests that walls of double-skin
steel construction with fibreglass or mineral wool insulation can achieve a four hour
fire rating. The siderails and fixings require special details which were included in
the test arrangements of the particular manufacturer and it is important that these
are followed closely. They include such things as providing slotted holes to allow
expansion of the rails rather than induce buckling, which may allow gaps to open
in the sheeting at joints.
References to Chapter 1
1. Horridge, J.F. (1985) Design of Industrial Buildings. Civil Engineering Steel
Supplement, November.
2. British Standards Institution (1998) Part 3: Code of practice for imposed roof
loads. BS 6399. BSI, London.
3. British Standards Institution (1997) Part 2: Code of practice for wind loads.
BS 6399. BSI, London.
4. British Standards Institution (2000) Part 1: Code of practice for design – Rolled
and welded sections. BS 5950. BSI, London.
5. King C.M. (2001) In-plane Stability of Portal Frames to BS 5950-1 : 2000.
(SCI-P292) The Steel Construction Institute, Ascot.
6. British Standards Institution (1998) Part 5: Code of practice for design of cold
formed thin gauge sections. BS 5950. BSI, London.
Steel Designers' Manual - 6th Edition (2003)
References
41
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
7. Davies J.M. & Raven G.K. (1986) Design of cold formed purlins. Thin
Walled Metal Structures in Buildings, pp. 151–60. IABSE Colloquium, Zurich,
Switzerland.
8. The Building Regulations (2002) Part L: Conservation of Fuel and Power. HMSO,
London.
Steel Designers' Manual - 6th Edition (2003)
Chapter 2
Multi-storey buildings
by ALAN HART and PHILIP PEACOCK
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
2.1 Introduction
The term multi-storey building encompasses a wide range of building forms. This
chapter reviews some of the factors that should be considered when designing
the type of multi-storey buildings commonly found in Europe, namely those less
than 15 storeys in height. Advice on designing taller buildings may be found in the
references to this chapter.1–5
2.1.1 The advantages of steel
In recent years the development of steel-framed buildings with composite metal
deck floors has transformed the construction of multi-storey buildings in the UK.
During this time, with the growth of increasingly sophisticated requirements for
building services, the very efficiency of the design has led to the steady decline
of the cost of the structure as a proportion of the overall cost of the building, yet
the choice of the structural system remains a key factor in the design of successful
buildings.
The principal reasons for the appeal of steel for multi-storey buildings are noted
below.
•
•
•
•
•
•
•
Steel frames are fast to erect.
The construction is lightweight, particularly in comparison with traditional
concrete construction.
The elements of the framework are prefabricated and manufactured under
controlled, factory conditions to established quality procedures.
The accuracy implicit in the manufacturing process by which the elements are
produced enables the designer to take a confident view of the geometric properties of the erected framework.
The dryness of the form of construction results in less on-site activities, plant,
materials and labour.
The framework is not susceptible to drying-out movement or delays due to slow
strength gain.
Steel frames have potential for adaptability inherent in their construction. Later
42
Steel Designers' Manual - 6th Edition (2003)
Introduction
•
43
modification to a building can be achieved relatively easily by unbolting a
connection; with traditional concrete construction such modifications would be
expensive, and more extensive and disruptive.
The use of steel makes possible the creation of large, column-free internal spaces
which can be divided by partitions and, by eliminating the external wall as a loadbearing element, allows the development of large window areas incorporated in
prefabricated cladding systems.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
2.1.2 Design aims
For the full potential of the advantages of steel-frame construction to be realized,
the design of multi-storey buildings requires a considered and disciplined approach
by the architects, engineers and contractors involved in the project. They must be
aware of the constraints imposed on the design programme by the lead time
between placing a contract for the supply of the steel frame and the erection of the
first pieces on site. The programme should include such critical dates on information release as are necessary to ensure that material order and fabrication can
progress smoothly.
The designer must recognize that the framework is the skeleton around which
every other element of the building will be constructed. The design encompasses
not only the structure but also the building envelope, services and internal finishes.
All these elements must be co-ordinated by a firm dimensional discipline, which
recognizes the modular nature of the components, to ensure maximum repetition
and standardization. Consequently it is impossible to consider the design of the
framework in isolation. It is vital to see the frame as part of an integrated building
design from the outset: the most efficient solution for the structure may not be effective in achieving a satisfactory solution for the total building.
In principle, the design aims can be considered under three headings:
•
•
•
Technical
Architectural
Financial.
Technical aims
The designer must ensure that the framework, its elements and connections are
strong enough to withstand the applied loads to which the framework will be subjected throughout its design life. The system chosen on this basis must be sufficiently
robust to prevent the progressive collapse of the building or a significant part of it
under accidental loading. This is the primary technical aim. However, as issues
related to strength have become better understood and techniques for the strength
design of frameworks have been formalized, designers have progressively used
Steel Designers' Manual - 6th Edition (2003)
44
Multi-storey buildings
lighter and stronger materials. This has generated a greater need to consider
serviceability, including dynamic floor response, as part of the development of the
structural concept.
Other important considerations are to ensure adequate resistance to fire and
corrosion. The design should aim to minimize the cost, requirements and intrusion
of the protection systems on the efficiency of the overall building.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Architectural aims
For the vast majority of buildings the most effective structural steel frame is the one
which is least obtrusive. In this way it imposes least constraint on internal planning,
and produces maximum usable floor area, particularly for open-plan offices. It also
provides minimal obstruction to the routeing of building services. This is an important consideration, particularly since building services are becoming more extensive
and demanding on space and hence on the building framework.
Occasionally the structure is an essential feature of the architectural expression
of the building. Under these circumstances the frame must achieve, among other
aims, a balance between internal planning efficiency and an expressed structural
form. However, these buildings are special, not appropriate to this manual, and will
not be considered in more detail, except to give a number of references.
Financial aims
The design of a steel frame should aim to achieve minimum overall cost. This is
a balance between the capital cost of the frame and the improved revenue from
early occupation of the building through fast erection of the steel frame: a more
expensive framework may be quicker to build and for certain uses would be more
economic to a client in overall terms. Commercial office developments are a good
example of this balance. Figure 2.1 shows a breakdown of construction costs for a
typical development.
2.1.3 Influences on overall design concept
Client brief
Clients specify their requirements through a brief. It is essential for effective design
to understand exactly the intentions of the client: the brief is the way in which the
client expresses and communicates these intentions. As far as the frame designer is
concerned, the factors which are most important are intended use, budget cost limits,
time to completion and quality. Once these are understood a realistic basis for producing the design will be established. The designer should recognize however that
in practice the brief is likely to evolve as the design develops.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Introduction
45
Fabric
A Demolition. Piling foundations and concrete work . . . . . . . . . . . . .11%
B Steel Frame. Deck and Fire Protection . . . . . . . . . . . . . . . . . . . . .10%
C Brickwork and Drywalling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4%
D External and internal Cladding and Sunscreens . . . . . . . . . . . . . .22%
E Roofing and Rooflights . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5%
52%
Finishes
F Ceiling and Floors . . . . . . . 7%
G Stone . . . . . . . . . . . . . . . . 5%
H Others . . . . . . . . . . . . . . . . 8%
20%
Fig. 2.1
Services
I Plumbing and Sprinklers . . .4%
J H.V.A.C. . . . . . . . . . . . . . .12%
K Electrics . . . . . . . . . . . . . . 8%
L Lifts . . . . . . . . . . . . . . . . . 4%
28%
Typical cost breakdown
Statutory constraints
The design of all buildings is subject to some form of statutory constraint. Multistorey buildings, particularly those in an urban environment, are subject to a high
level of constraint, which will generally be included in the conditions attached to
the granting of outline planning permission. The form and degree that this may take
could have significant impact on the frame design. For example, street patterns
and lighting restrictions may result in a non-rectilinear plan with a ‘stepped-back’
structure. If an appropriate layout is to be provided it is vital to understand these
constraints from the outset.
Certain government buildings, financial headquarters and other strategically
sensitive buildings may need to be designed to resist terrorist threats. Provision may
be in the form of blast-resistant flows, walls and facades to vulnerable areas.
Physical factors
The building must be designed to suit the parameters determined by its intended
use and its local environment.
Steel Designers' Manual - 6th Edition (2003)
46
Multi-storey buildings
Its intended use will dictate the intensity of imposed loadings, the fire protection
and corrosion resistance requirements and the scope of building services.
Certain government buildings, financial headquarters and other strategically sensitive buildings may need to be designed to resist terrorist threats. Provision may
be in the form of blast-resistant floors, walls and façades to vulnerable areas.
The local environment will dictate the lateral load requirements but more importantly it will determine the nature of the existing ground conditions. To achieve
overall structural efficiency it is essential that the structural layout of the frame is
responsive to the constraints imposed by these ground conditions.
These factors are considered in more detail in the next section.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
2.2 Factors influencing choice of form
Environmental
There are a number of factors which influence the choice of structural form that are
particular to the site location. These can have a dominant effect on the framing
arrangement for the structure.
The most obvious site-dependent factors are related to the ground conditions.
A steel-framed building is likely to be about 60% of the weight of a comparable
reinforced concrete building. This difference will result in smaller foundations with
a consequent reduction on costs. In some cases this difference in weight enables
simple pad foundations to be used for the steel frame where the equivalent
reinforced concrete building would require a more complex and expensive solution.
For non-uniformly loaded structures it will also reduce the magnitude of differential settlements and for heavily loaded structures may make possible the use of a
simple raft foundation in preference to a large capacity piled solution (Fig. 2.2).
Difficult ground conditions may dictate the column grid. Long spans may be
required to bridge obstructions in the ground. Such obstructions could include, for
example, buried services, underground railways or archaeological remains. Generally, a widely spaced column grid is desirable since it reduces the number of foundations and increases the simplicity of construction in the ground.
Other site-dependent constraints are more subtle. In urban areas they relate to
the physical constraints offered by the surrounding street plan, and the rights of
light of adjoining owners. They also relate to the planning and architectural objectives for specific sites. The rights of light issues or planning considerations may
dictate that upper floors are set back from the perimeter resulting in stepped construction of the upper levels. Invariably the resulting framing plan is not rectilinear
and may have skew grids, cantilevers and re-entrant corners.
These constraints need to be identified early in the design in order that they are
accommodated efficiently into the framing. For example, wherever possible,
stepped-back façades should be arranged so that steps take place on the column
grid and hence avoid the need for heavy bridging structures. In other situations the
designer should always investigate ways in which the impact of lack of uniformity
in building form can be contained within a simple structural framing system which
generates a minimum of element variations and produces simple detailing.
Steel Designers' Manual - 6th Edition (2003)
Factors influencing choice of form
47
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
20% reduction
approximate
rot t foundation
Fig. 2.2
piled foundation
Foundation savings
Building use
The building use will dictate the planning module of the building, which will in turn
determine the span and column grids. Typical grids may be based on a planning
module of 600/1200 mm or 500/1500 mm. However, the use has much wider impact,
particularly on floor loadings and building services. The structural arrangement, and
depth selected, must satisfy and accommodate these requirements.
For example, financial-dealing floors require clear open spaces located on the
lower floors, which would dictate a different structural solution to the rest of the
building.This may necessitate the use of a transfer structure to carry the upper floors
on an economical column grid (Fig. 2.3).
Floor loadings
Because steel-framed buildings are relatively light in weight, excessive imposed
loadings will have a greater effect on the sizing of structural components, particularly floor beams, than with reinforced concrete structures.
The floor loadings to be supported by the structure have two components:
•
The permanent or dead loading comprising the self-weight of the flooring and
the supporting structure together with the weight of finishes, raised flooring,
ceiling, air-conditioning ducts and equipment.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
48
Multi-storey buildings
Fig. 2.3
Typical load transfer systems
•
The imposed loading, which is the load that the floor is likely to sustain during
its life and which will depend on the building use. Imposed floor loads for various
types of building are governed by BS 6399 but the standard loading for office
buildings is usually 4 kN/m2 with an additional allowance of 1 kN/m2 for movable
partitioning.
For normal office loadings, dead and imposed loadings are roughly equal in
proportion but higher imposed load allowances will be necessary in plantrooms or
to accommodate special requirements such as storage or heavy equipment.
Floor beams will be designed to limit deflection under the imposed loadings.
British Standard BS 5950 governing the design of structural steelwork sets a limit
for deflection under imposed loading of (span/200) generally and (span/360) where
there are brittle finishes. Edge beams supporting cladding will be subject to restriction on deflection of 10–15 mm. Deflections may be noticeable in the ceiling layout
and should be taken into account when determining the available clearance for
service routes. The designer should therefore check the cumulative effect of deflections in the individual members of a floor system although the actual maximum displacement is in practice almost always less than that predicted. In some instances,
vibrations of floor components may cause discomfort or affect sensitive equipment,
and the designer should check the fundamental response of the floor system. The
threshold of perceptible vibrations in building is difficult to define, and present limits
are rather arbitrary. There is some evidence that modern lightweight floors can
be sensitive to dynamic loads, which may have an effect on delicate equipment.
However, in most situations a simple check on the natural frequency of the floor
system is all that is required.
Building services and finishes
In buildings requiring anything other than minimal electrical services distribution,
the inter-relationship of the structure, the mechanical and electrical services and the
building finishes will need to be considered together from the outset.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Factors influencing choice of form
49
It is essential to co-ordinate the details of the building services, cladding and structure at an early stage of the project in order to produce a building which is simple
to fabricate and quick to erect. Apparently minor variations to the steelwork,
brought about by services and finishes requirements defined after a steel fabrication contract has been let, can have a disproportionate effect on the progress of
fabrication and erection. Steel buildings impose a strict discipline on the designer
in terms of the early production of final design information. If the designer fails to
recognize this, the advantages of steel-framed building cannot be realized.
The integration of the building services with the structure is an important factor
in the choice of an economic structural floor system. The overall depth of the floor
construction will depend on the type and distribution of services in the ceiling void.
The designer may choose to separate the structural and services zones or accommodate the services by integrating them with the structure, allowing for the structural system to occupy the full depth of the floor construction. (See Fig. 2.4.)
Separation of zones usually requires confining the ducts, pipes and cables to a
horizontal plane below the structure, resulting in either a relatively deep overall
floor construction or close column spacings. Integration of services with structure
requires either deep perforated structural components or vertical zoning of the
services and structure.
For the range of structural grids used in conventional building, traditional steel
floor construction is generally deeper than the equivalent reinforced concrete flat
slab: the difference is generally 100–200 mm for floor structures which utilize composite action and greater for non-composite floors (Fig. 2.5). The increased depth is
only at the beam position; elsewhere, between beams, the depth is much less and
the space between them may accommodate services, particularly if the beams may
be penetrated (Fig. 2.6). The greater depth of steel construction does not therefore
necessarily result in an increase in building height if the services are integrated
within the zone occupied by the structure. A number of possible solutions exist for
raised floor
raised floor
slob
slab
structure
7
+I
services zone
I+
(b)
structure
and services
ceiling
ceiling
Fig. 2.4 Building services and floor structure: (a) separation of services and structure; (b)
integration of services and structure
Steel Designers' Manual - 6th Edition (2003)
50
Multi-storey buildings
(b)
(c)
> 200 mm
— —
Fig. 2.5
Overall floor depths: (a) R.C. flat slab; (b) composite; (c) non-composite
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(a)
(b)
Fig. 2.6 Ceiling voids: (a) steel frame: variable void height; (b) concrete slab: constant void
height
integrated systems, particularly in long-span structures utilizing castellated, cellular
or stub-girder beams. (See Fig. 2.7.)
A number of solutions have been developed which allow long spans to co-exist
with separation of the building services by profiling the steel beam to provide space
for services, either at the support or in the span. Automated plate cutting and
welding techniques are used to produce economical profiled plate girders, with or
without web openings. (See Fig. 2.8.)
Overall depth may be reduced by utilizing continuous or semi-continuous rather
than simple connections at the ends of the beams. This reduces the maximum
bending moment and deflection. However, such solutions are not as efficient as
would first appear since the non-composite section at the support is much less
efficient than the composite section at mid-span. Indeed, if the support bending
moments are large in comparison with the span bending moments the depth
may be greater than the simply-supported composite beam. This is an expensive
fabrication in comparison with straight rolled beam sections. (See Fig. 2.9.)
In addition, the use of continuous joints can increase column sizes considerably.
Semi-continuous braced frames can provide an economic balance between the
primary benefits associated with simple or continuous design alternatives. The
degree of continuity between the beams and columns can be chosen so that complex
stiffening to the column is not required. Methods of analysis have been developed
for non-composite construction to permit calculation by hand. It is possible to
achieve reduced beam depths and reduced beam weights.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Factors influencing choice of form
(a)
(b)
(c)
(d(
51
Fig. 2.7 Integration of services: (a) separated (traditional); (b) integrated (shallow floor
‘Slimdek®’ system); (c) integrated (long span ‘primary’ beam – stub girder); (d) integrated (long span ‘secondary’ beams)
potentol duct zone
taper beam -
VAV box
duct
ceiling grid
duct
zone
(extract)
Fig. 2.8
duct
zone
(supply)
Tapered beams and services
zone
(supply)
Steel Designers' Manual - 6th Edition (2003)
Multi-storey buildings
52
01
02<01
deflection
-H h-
4-
4-
Ms2 moment of
resistance
of connection
-I-
T
bending moment = —— - Ms 2
bending moment = ——
lb)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(a)
03<02
RH
Ms3> Ms21
T
cl
Fig. 2.9
Floor depth: (a) simple; (b) semi-continuous; (c) continuous
The overall depth may also be reduced by using higher-strength steel, but this is
only of advantage where the element design is controlled by strength. The stiffness
characteristics of both steels are the same: hence, where deflection or vibration
govern, no advantage is gained by using the stronger steel.
Recently, shallow floor systems have been developed for spans up to about 9 m
which allow integration of services within the slab depth. Structural systems range
from conventional fabricated beams using precast units to proprietary systems using
new asymmetric rolled beams and deep metal decking. These approaches can form
the basis of energy-saving sustainable solutions.
Semi-continuous braced frames can provide an economic balance between the
primary benefits associated with simple or continuous design alternatives. The
degree of continuity between the beams and columns can be chosen so that complex
stiffening to the column is not required. Methods of analysis have been developed
for non-composite construction to permit calculation by hand. It is possible to
achieve reduced beam depths and reduced beam weights.
Steel Designers' Manual - 6th Edition (2003)
Factors influencing choice of form
53
External wall construction
The external skin of a multi-storey building is supported off the structural frame.
In most high quality commercial buildings the cost of external cladding systems
greatly exceeds the cost of the structure. This influences the design and construction of the structural system in a number of ways:
•
•
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
•
•
The perimeter structure must provide a satisfactory platform to support the
cladding system and be sufficiently rigid to limit deflections of the external wall.
A reduction to the floor zone may significantly reduce the area and hence cost
of cladding.
Fixings to the structure should facilitate rapid erection of cladding panels.
A reduction in the weight of cladding at the expense of cladding cost will not
necessarily lead to a lower overall construction cost.
Lateral stiffness
Steel buildings must have sufficient lateral stiffness and strength to resist wind and
other lateral loads. In tall buildings the means of providing sufficient lateral stiffness forms the dominant design consideration. This is not the case for low- to
medium-rise buildings.
Most multi-storey buildings are designed on the basis that wind and/or notional
horizontal forces acting on the external cladding are transmitted to the floors, which
form horizontal diaphragms transferring the lateral load to rigid elements and then
to the ground. These rigid elements are usually either braced-bay frames, rigidjointed frames, reinforced concrete or steel–concrete–steel sandwich shear walls.
Low-rise unbraced frames up to about six storeys may be designed using the simplified wind-moment method. In this design procedure, the frame is made statically
determinate by treating the connections as pinned under vertical load and rigid
under horizontal loads. This approach can be used on both composite and noncomposite frames, albeit with strict limitations on frame geometry, loading patterns
and member classification.
British Standard BS 5950 sets a limit on lateral deflection of columns as height/300
but height/600 may be a more reasonable figure for buildings where the external
envelope consists of sensitive or brittle materials such as stone facings.
Accidental loading
A series of incidents in the 1960s culminating in the partial collapse of a systembuilt tower block at Ronan Point in 1968 led to a fundamental reappraisal of the
approach to structural stability in building.
Traditional load-bearing masonry buildings have many in-built elements providing inherent stability which are lacking in modern steel-framed buildings. Modern
structures can be refined to a degree where they can resist the horizontal and
Steel Designers' Manual - 6th Edition (2003)
54
Multi-storey buildings
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
vertical design loadings with the required factor of safety but may lack the ability
to cope with the unexpected.
It is this concern with the safety of the occupants and the need to limit the extent
of any damage in the event of unforeseen or accidental loadings that has led to the
concept of robustness in building design. Any element in the structure that supports
a major part of the building either must be designed for blast loading or must be
capable of being supported by an alternative load path. In addition, suitable ties
should be incorporated in the horizontal direction in the floors and in the vertical
direction through the columns. The designer should be aware of the consequences
of the sudden removal of key elements of the structure and ensure that such an
event does not lead to the progressive collapse of the building or a substantial part
of it. In practice, most modern steel structures can be shown to be adequate without
any modification.
Cost considerations
The time taken to realize a steel building from concept to completion is generally
less than that for a reinforced concrete alternative. This reduces time-related building costs, enables the building to be used earlier, and produces an earlier return on
the capital invested.
To gain full benefit from the ‘factory’ process and particularly the advantages
of speed of construction, prefabrication, accuracy and lightness, the cladding and
finishes of the building must have similar attributes. The use of heavy, slow and in
situ finishing materials is not compatible with the lightweight, prefabricated and fast
construction of a steel framework.
The cost of steel frameworks is governed to a great extent by the degree of simplicity and repetition embodied in the frame components and connections. This also
applies to the other elements which complete the building.
The criterion for the choice of an economic structural system will not necessarily be to use the minimum weight of structural steel. Material costs represent only
30–40% of the total cost of structural steelwork.The remaining 60–70% is accounted
for in the design, detailing, fabrication, erection and protection. Hence a choice
which needs a larger steel section to avoid, say, plate stiffeners around holes or
allows greater standardization will reduce fabrication costs and may result in the
most economic overall system.
Because a steel framework is made up of prefabricated components produced in
a factory, repetition of dimensions, shapes and details will streamline the manufacturing process and is a major factor in economic design (Fig. 2.10).
Fabrication
The choice of structural form and method of connection detailing have a significant
impact on the cost and speed of fabrication and erection. Simple braced frameworks
with bolted connections are considered the most economic and the fastest to build
for low- to medium-rise buildings.
Steel Designers' Manual - 6th Edition (2003)
Factors influencing choice of form
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 2.10
55
Structural costs: (a) economic and (b) uneconomic layouts
Economy is generally linked to the use of standard rolled sections but, with the
advent of automated cutting and welding equipment, special fabricated sections are
becoming economic if there is sufficient repetition.
The development of efficient, automated, cold-sawing techniques and punching
and drilling machines has led to the fabrication of building frameworks with bolted
assemblies. Welded connections involve a greater amount of handling in the fabrication shop, with consequent increases in labour and cost.
Site-welded connections require special access, weather protection, inspection
and temporary erection supports. By comparison, on-site bolted connections enable
the components to be erected rapidly and simply into the frame and require no
further handling.
The total weight of steel used in continuous frames is less than in semicontinuous or simple frames, but the connections for continuous frames are more
complex and costly to fabricate and erect. On balance, the cost of a continuous frame
structure is greater, but there may be other considerations which offset this cost
differential. For example, in general the overall structural depth of continuous
frames is less. This may reduce the height of the building or improve the distribution
of building services, both of which could reduce the overall cost of the building.
Corrosion protection to internal building elements is an expensive and timeconsuming activity. Experience has shown that it is unnecessary for most internal
locations and consequently only steelwork in risk areas should require any protection. Factory-applied coatings of intumescent fire protection can be cost-effective
and time-saving by removing the operation from the critical path.
Construction
A period of around 8–12 weeks is usual between placing a steel order and the arrival
of the first steel components on site. Site preparation and foundation construction
generally take a similar or longer period (see Fig. 2.11). Hence, by progressing fabrication in parallel with site preparation, significant on-site construction time may
be saved, as commencement of shop fabrication is equivalent to start-on-site for an
in situ concrete-framed building. By manufacturing the frame in a factory, the risks
'
U
'
—
.
—
—
—
'
U
'
•
U
—
—
'
'
—
IT
•
•
Typical progress schedule (in weeks)
—
—
—
—
121620242832364044485256606468727680848892961001041088
o
a)
0
-Fig.
2.11
U-
E
0
—
—
—
4
EXTERNAL
SERVICES
COMMISSIONING
& WORKS
DECKING
STEELWORK ERECT
COMPOSITE&
____________________________________________
& Basement Lower Construct
Piling Bored
piling for Prepare & Excavate
FOUNDATIONS CONSTRUCT & EXCAVATE
Foundations
(Non-critical) Floor Ground Bays Perimeter
(Non-critical) Drains & Basement Construct
SLAB GROUND & DRAINS UNDERGROUND
SLABS
FIRE
PANELS CLADDING
& Sunscreens & External
site) on (sprayed PROOFING
Atrium Internal
WATERPROOFING
Balconies and Roof
FINISHI INTERNAL & PARTITIONS BRICKWORK
GRANITE
SITE CLEAR
56
protection fire off-site Eusing
time-saving possible weeks 4
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
Multi-storey buildings
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Factors influencing choice of form
57
of delay caused by bad weather or insufficient or inadequate construction resources
in the locality of the site are significantly reduced.
Structural steel frameworks should generally be capable of being erected without
temporary propping or scaffolding, although temporary bracing will be required,
especially for welded frames. This applies particularly to the construction of the concrete slab, which should be self-supporting at all stages of erection. Permanent metal
or precast concrete shutters should be used to support the in situ concrete.
In order to allow a rapid start to construction, the structural steelwork frame
should commence at foundation level, and preference should be given to single
foundations for each column rather than raft or shared foundations (Fig. 2.12).
Speed of erection is directly linked to the number of crane hours available. To
reduce the number of lifts required on site, the number of elements forming the
framework should be minimized within the lifting capacity of the craneage provided
on site for other building components. For similar-sized buildings, the one with the
longer spans and fewer elements will be the fastest to erect. However, as has been
mentioned earlier, longer spans require deeper, heavier elements, which will
increase the cost of raw materials and pose a greater obstruction to the distribution
of building services, thereby requiring the element to be perforated or shaped and
hence increasing the cost of fabrication.
Columns are generally erected in multi-storey lengths: two is common and three
is not unusual. The limitation on longer lengths is related more to erection than to
restrictions on transportation, although for some urban locations length is a major
consideration for accessibility.
To provide rapid access to the framework the staircases should follow the erection of the frame. This is generally achieved by using prefabricated stairs which are
detailed as part of the steel frame.
The speed of installation of the following building elements is hastened if their
connection and fixing details are considered at the same time as the structural steel
frame design. In this way the details can be either incorporated in the framework
or separated from it, whichever is the most effective overall: it is generally more
efficient to separate the fixings and utilize the high inherent accuracy of the frame
:'
n
11
n
U__[[I
a__U__[[I
Li
Fig. 2.12
Columns on large diameter bored piles
Steel Designers' Manual - 6th Edition (2003)
58
Multi-storey buildings
to use simple post-fixed details, provided these do not require staging or scaffolding to give access.
Finally, on-site painting extends the construction period and provides potential
compatibility problems with following applied fire protection systems. Painting
should therefore only be specified when absolutely necessary.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
2.3 Anatomy of structure
In simple terms, the vertical load-carrying structure of a multi-storey building
comprises a system of vertical column elements interconnected by horizontal beam
elements which support floor-element assemblies. The resistance to lateral loads is
provided by diagonal bracing elements, or wall elements, introduced into the vertical rectangular panels bounded by the columns and beams to form vertical trusses,
or walls. Alternatively, lateral resistance may be provided by developing a continuous or semi-continuous frame action between the beams and columns. The flexibility of connections should be taken into account in the analysis. All structures should
have sufficient sway stiffness, so that the vertical loads acting with lateral displacements of the structure do not induce excessive secondary forces or moments in the
members or connections. A building frame may be classed as ‘non-sway’ if the sway
deformation is sufficiently small for the resulting secondary forces and moments
to be negligible. In all other cases the building frame should be classed as ‘swaysensitive’. The stiffening effect of cladding and infill wall panels may be taken into
account by using the method of partial sway bracing. The floor-element assemblies
provide the resistance to lateral loads in the horizontal plane.
In summary, the components of a building structure are columns, beams, floors
and bracing systems (Fig. 2.13).
2.3.1
Columns
These are generally standard, universal column, hot-rolled sections. They provide a
compact, efficient section for normal building storey heights. Also, because of the
section shape, they give unobstructed access for beam connections to either the
flange or web. For a given overall width and depth of section, there is a range of
weights which enable the overall dimensions of structural components to be nominally maintained for a range of loading intensities.
Where the loading requirements exceed the capacity of standard sections, additional plates may be welded to the section to form plated columns, or fabricated
columns may be formed by welding plates together to form a plate-column (Fig. 2.14).
The use of circular or rectangular tubular elements marginally improves the loadcarrying efficiency of components as a result of their higher stiffness-to-weight ratio.
Concrete filling significantly improves the axial load-carrying capacity and fire
resistance.
Steel Designers' Manual - 6th Edition (2003)
Anatomy of structure
59
primary beam
floor
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
bracing
column
Fig. 2.13
Conventional steel frame components
(a)
(b)
(d)
p
Fig. 2.14 Types of column: (a) plated (by addition of plates to U.C. section); (b) universal;
(c) tubular; (d) fabricated plate
Steel Designers' Manual - 6th Edition (2003)
60
2.3.2
Multi-storey buildings
Beams
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Structural steel floor systems consist of prefabricated standard components, and
columns should be laid out on a repetitive grid which establishes a standard structural bay. For most multi-storey buildings, functional requirements will determine
the column grid which will dictate spans where the limiting criterion will be stiffness rather than strength (Fig. 2.15).
Steel components are uni-directional and consequently orthogonal structural
column and beam grids have been found to be the most efficient. The most efficient
floor plan is rectangular, not square, in which main, or ‘primary’, beams span the
shorter distance between columns and closely-spaced ‘secondary’ floor beams span
the longer distance between main beams. The spacing of the floor beams is controlled by the spanning capability of the concrete floor construction (Fig. 2.16).
Fig. 2.15
Typical floor layout
Steel Designers' Manual - 6th Edition (2003)
Anatomy of structure
I
I
I
——
I
61
I
1
primary beams
secondary beams
I
I
I
I
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
2.4-3.0 m spocing
(a)
(b)
Fig. 2.16
Beam and shallow deck layout: (a) inefficient; (b) efficient
Having decided on the structural grid, the designer must choose an economic
structural system to satisfy all the design constraints. The choice of system and its
depth depends on the span of the floor (Fig. 2.17). The minimum depth is fixed by
practical considerations such as fitting practical connections. As the span increases,
the depth will be determined by the bending strength of the member and, for longer
spans, by the stiffness necessary to prevent excessive deflection under imposed load
or excessive sensitivity to induced vibrations (Fig. 2.18). For spans up to 9 m, shallow
beams with precast floors or deep composite metal deck floors can be used to
.
Composite trusses
I
Tapered fabricated beam
I
Stub girder
I
Castellated or cellular beams and shallow metal deck
Plain composite beams and shallow metal deck
Simple beams with precast slabs
Slim floor and deep metal deck
5
I
I
10
15
Beam span (m)
Fig. 2.17
Span ranges for different composite floor systems
20
Steel Designers' Manual - 6th Edition (2003)
Multi-storey buildings
62
Vibration
1000
E
E
Deflection
0
=
=
Strength
C-)
500
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Minimum
0
10
5
15
20
Beam span (m)
Fig. 2.18
Structural criteria governing choice
minimize the floor zone. Simple universal beams with precast floors or composite
metal deck floors are likely to be most economic for spans up to 12 m. A range of
section capacities for each depth enables a constant depth of construction to be
maintained for a range of spans and loading. As with column components, plated
beams and fabricated girders may be used for spans above 10–12 m. They are particularly appropriate where heavier loading is required and overall depth is limited.
For medium to lightly loaded floors and long spans, beams may also take the form
of castellated beams fabricated from standard sections, cellular sections or plates.
Above 15 m, composite steel trusses may be economic. As the span increases, the
depth and weight of the structural floor increase, and above 15 m spans depth predominates because of the need to achieve adequate stiffness.
Castellated and cellular beam sections
Castellated beams (Fig. 2.19(a)) have been used for many years to increase the
bending capacity of the beam section and to provide limited openings for services.
These openings are rarely of sufficient size for ducts to penetrate without significant modification, which increases fabrication cost. The cellular concept is a
development of castellated beams that provides circular openings and greater shear
capacity. Since their introduction in 1990, they have proved to be increasingly
popular for longer span solutions where services and structure have to be integrated.
Bespoke openings for services can be cut in the webs of universal beams and
fabricated plate girders.
Steel Designers' Manual - 6th Edition (2003)
—
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(a) castellated beam
Anatomy of structure
63
(b) cellular beam
C
(
(c) beam with web openings
Fig. 2.19
Beams with web openings for service penetrations
Fabricated plate girders
Conventional universal beams span a maximum of about 15 m. Recent advances in
automatic and semi-automatic fabrication techniques have allowed the economic
production of plate girders for longer span floors. Particularly if a non-symmetric
plate girder is used, it is possible to achieve economic construction well in excess of
15 m (Fig. 2.20). Such plate girders can readily accommodate large openings for
major services. If these are in regions of high stress, single-sided web stiffening may
be used. Away from regions of high stress, stiffening is usually not required.
The use of intumescent paints, applied offsite, is becoming increasingly popular.
One major fabricator is now offering an integrated design and fabrication service
for customized plate girders which can achieve a fire resistance of 2 hours when
applied as a single layer in an off-site, factory-controlled process.
Taper beams
Taper beams (Fig. 2.20) are similar to fabricated plate girders except that their depth
varies from a maximum in mid-span to a minimum at supports, thus achieving a
highly efficient structural configuration. For simply-supported composite taper
beams in buildings the integration of the services can be accommodated by locating the main ducts close to the columns. Alternative taper beam configurations can
be used to optimize the integration of the building services.
Steel Designers' Manual - 6th Edition (2003)
64
Multi-storey buildings
IVZf/Z f/4
equal flanges
V,W/ 7Z7W4
1wJ /f//p
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
unequal flanges
fish belly
linear taper
horizontal/taper
haunched taper
optimum spans 10—18 m
Fig. 2.20
Fabricated plate girders and taper beams
Composite steel floor trusses
Use of composite steel floor trusses as primary beams in the structural floor system
permits much longer spans than would be possible with conventional universal
beams.The use of steel trusses for flooring systems is common for multi-storey buildings in North America but seldom is used in Britain. Although they are considerably lighter than the equivalent universal beam section the cost of fabrication is very
Steel Designers' Manual - 6th Edition (2003)
Anatomy of structure
65
AJ
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Vierendeel panel
A
Fig. 2.21
Composite truss
much greater, as is the cost of fireproofing the truss members. For maximum
economy, trusses should be fabricated from T-sections and angles using simple
welded lap joints. The openings between the diagonal members should be designed
to accept service ducts, and if a larger opening is required a Vierendeel panel can
be incorporated at the centre of the span. Because a greater depth is required for
floor trusses, the integration of the services is always within the structural zone (Fig.
2.21).
Stub girder construction
Stub girders were developed in North America in the 1970s as an alternative form
of construction for intermediate range spans of between 10 and 14 m. They have not
been used significantly in the UK. Figure 2.22 shows a typical stub girder with a
bottom chord consisting of a compact universal column section which supports the
secondary beams at approximately 3-metre centres. Between the secondary beams
a steel stub is welded on to the bottom chord to provide additional continuity and
to support the floor slab. The whole system acts as a composite Vierendeel truss. A
disadvantage of stub girders is that the construction needs to be propped while the
concrete is poured and develops strength. Arguably, a deep universal beam with
large openings provides a more cost-effective alternative to the stub girder because
of the latter’s high fabrication content.
Steel Designers' Manual - 6th Edition (2003)
Multi-storey buildings
66
3000
3000
•
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
bottom
chord
:
3000
3000
stub
slob
stub
bottom chord
_______ (universal column)
Fig. 2.22
Stub girder
2.3.3 Floors
These take the form of concrete slabs of various forms of construction spanning
between steel floor beams (Fig. 2.23). The types generally found are:
•
•
•
•
in situ concrete slab cast on to permanent profiled shallow or deep metal decking,
acting compositely with the steel floor beams;
precast concrete slabs acting non-compositely with the floor beams:
in situ concrete slab, with conventional removable shuttering, acting compositely
with the floor beams;
in situ concrete slab cast on thin precast concrete slabs to form a composite slab,
which in turn acts compositely with the floor beams.
The most widely used construction internationally is profiled shallow metal decking.
Composite action with the steel beam is normally provided by shear connectors
welded through the metal decking on to the beam flange. Shallow floor systems
using deep metal decking are gaining popularity in the UK although precast concrete systems are still used extensively. Composite action enables the floor slab to
work with the beam, enhancing its strength and reducing deflection (Fig. 2.24).
Because composite action works by allowing the slab to act as the compression
flange of the combined steel and concrete system, the advantage is greatest when
the beam is sagging. Consequently composite floor systems are usually designed as
simply supported.
Steel Designers' Manual - 6th Edition (2003)
Anatomy of structure
67
grouted joint
unit
(a)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
///////7 —
precast
—L
(b)
shallow
metal
deck
concrete
shutter
(d)
(C)
welded shear stud
,— asymetric slimfior'
J beam (ASB)
I—
•1
(e)
deep
metal
deck
k—deep
metal
deck
Fig. 2.23 Floor construction: (a) precast (non-composite); (b) in situ (composite); (c) in
situ/precast (composite); (d) in situ/shallow metal decking (composite); (e) Slimfloor – in situ/deep metal decking (composite); (f) Slimdek® – in situ/deep metal
decking (composite)
Fig. 2.24
Metal deck floor slabs
Shallow metal deck floor construction
Experience has shown that the most efficient floor arrangements are those using
shallow metal decking spanning about 3–4.5 m between floor beams. For these spans
the metal decking does not normally require propping during concreting and the
Steel Designers' Manual - 6th Edition (2003)
Multi-storey buildings
68
concrete thicknesses are near the practical minimum for consideration of strength
and fire separation.
Steel studs are welded through the decking on the flange of the beams below to
form a connection between steel beam and concrete slab. Concrete, which may be
either lightweight or normal weight, is then poured on to the decking, usually by
pumping, to make up the composite system. Shallow metal decking acts both as permanent formwork for the concrete and as tensile reinforcement for the slab. There
are many types of steel decking available (Fig. 2.25(a)) but perhaps the most commonly used is the re-entrant profile type, which provides a flat soffit and facilitates
fixings for building services and ceilings.
Some of the advantages of composite shallow metal deck floor construction are:
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
•
•
•
•
•
•
Steel decking acts as permanent shuttering, which can eliminate the need for slab
reinforcement and, due to its high stiffness and strength, propping of the construction while the wet concrete develops strength.
Composite action reduces the overall depth of structure.
It provides up to 2 hours fire resistance without additional fire protection and 4
hours with added thickness or extra surface protection.
It is a light, adaptable system that can be easily manhandled on site, cut to
awkward shapes and drilled or cut out for additional service requirements.
Lightweight construction reduces frame loadings and foundation costs.
It allows simple, rapid construction techniques.
Figure 2.26 illustrates alternative arrangements of primary and secondary beams for
a deck span of 3 m.
(a)
Fig. 2.25
(b)
Metal deck profiles: (a) shallow deck (50–100 mm); (b) deep deck (150–250 mm)
Steel Designers' Manual - 6th Edition (2003)
Anatomy of structure
69
11111:1 Jooo
99 3OOO
(a)
(b)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 2.26 Alternative framing systems for floors: (a) long span secondary beams; (b) long
span primary beams
Deep metal deck, shallow floor construction
Deep metal decks are normally used to create shallow floors e.g. Slimflor® and
Slimdek® construction. The deep metal deck extends the span capability up to 9 m;
however, the deck and/or support beams may require propping during concreting.
An additional tensile reinforcement bar is provided within the ribs of the deep
decking to improve the load carrying capacity and fire resistance of the floor slab.
Although there are several variants internationally, Slimflor® construction in the UK
comprises a universal column with a plate welded to the underside supporting the
deep metal decking. Shear studs are shop-welded to the top flange of the beam to
form a connection between the steel beam and concrete slab. Slimdek® construction is a technologically advanced solution comprising a rolled asymmetric beam
(ASB) which supports the deep metal decking directly. Shear connection is achieved
through the bond developed between the steel beam and concrete encasure. These
features reduce material and fabrication content. Partial concrete encasement of the
steel beam provides up to 1 hour inherent fire resistance.
The range of deep metal deck profiles (Fig. 2.25(b)) is more limited than for
shallow decks, and those available carry similar attributes and advantages. Some
additional advantages of Slimflor® and Slimdek® construction are:
•
•
•
•
•
The shallow composite slab achieves excellent load capacity, diaphragm action
and robustness
There are fewer frame components to erect, saving construction time
The shallow floor construction allows more floors for a given building height or
reduces the cost of cladding and vertical services, lift shafts, etc.
It provides up to 1 hour inherent fire resistance and up to 2 hours using passive
fire protection to the beam soffit only
Large openings for vertical services can be formed in the floor slab without the
need for secondary framing
Steel Designers' Manual - 6th Edition (2003)
Multi-storey buildings
70
•
•
•
•
Services can be integrated between the decking ribs passing through webopenings in the beam
ASBs achieve composite action without the need for shear studs
Rectangular hollow section edge beams provide good torsional resistance and
maintain the shallow floor depth
Floor slabs can be used for fabric energy storage forming part of an environmentally sustainable building solution.
Figure 2.27 illustrates a typical beam layout at the building perimeter.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Precast floor systems
Universal beams supporting precast prestressed floor units (Fig. 2.28) have some
advantages over other forms of construction. Although of heavier construction
than comparable composite metal deck floors, this system offers the following
advantages.
•
•
•
•
Fewer floor beams since precast floor units can span up to 6–8 m without
difficulty.
No propping is required.
Shallow floor construction can be obtained by supporting precast floor units on
shelf angles or on wide plates attached to the bottom flanges of universal columns
acting as beams (Slimfloor).
Fast construction because no time is needed for curing and the development of
concrete strength.
,i.
Ii
ASB
'1 deep
deck
5 to 9 m
fr span
——I
RHSFB edge beam
5 to 9 m
Fig. 2.27
Slimdek® floor arrangement
-Ff1 column (UC or SHSI
tie
member
Steel Designers' Manual - 6th Edition (2003)
Anatomy of structure
6000
71
6000
H—
/
_T
9o00
/i' /:-
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
elements
precast floor
elements
.
•..
ceiling line
Fig. 2.28
Precast concrete floors
On the other hand the disadvantages are:
•
•
Composite and diaphragm action is not readily achieved without a structural
floor screed.
Heavy floor units are difficult to erect in many locations and require the use of
a tower crane, which may have implications for the construction programme.
2.3.4 Bracings
Three structural systems are used to resist lateral loads: continuous or wind-moment
frames, reinforced concrete walls and braced-bay frames (Fig. 2.29). Combinations
of these systems may also be used.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
72
Multi-storey buildings
/\
U
U
U
El
><
><
><
><
eccentric
cross
K
Fig. 2.29 Bracing structures: (a) continuous frame; (b) reinforced concrete wall; (c) braced
bay frames
Continuous construction
Continuous frames are those with rigid moment-resisting connections between
beam and columns. It is not necessary that all connections in a building are detailed
in this way: only sufficient frames to satisfy the performance requirements of the
building.
The advantage of a continuous frame is:
•
Provides total internal adaptability with no bracings between columns or walls
to obstruct circulation.
Steel Designers' Manual - 6th Edition (2003)
Anatomy of structure
73
However, the disadvantages are:
•
•
•
•
Increased fabrication for complex framing connections
Increased site connection work, particularly if connections are welded
Columns are larger to resist bending moments
Generally, less stiff than other bracing systems.
Wind-moment frames are limited in application.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Shear walls
Reinforced concrete walls constructed to enclose lift, stair and service cores generally possess sufficient strength and stiffness to resist the lateral loading.
Cores should be located to avoid eccentricity between the line of action of the
lateral load and the centre of stiffness of the core arrangement. However, the core
locations are not always ideal because they may be irregularly shaped, located at
one end of the building or are too small. In these circumstances, additional braced
bays or continuous frames should be provided at other locations (Fig. 2.30).
Although shear walls have traditionally been constructed in in situ reinforced
concrete they may also be constructed of either precast concrete or brickwork.
additional bracing/rigid
frame required
Fig. 2.30
Core locations: (a) efficient; (b) inefficient
Steel Designers' Manual - 6th Edition (2003)
74
Multi-storey buildings
The advantages of shear walls are:
•
•
•
•
The beam-to-column connections throughout the frame are simple, easily fabricated and rapidly erected.
Shear walls tend to be thinner than other bracing systems and hence save space
in congested areas such as service and lift cores.
They are very rigid and highly effective.
They act as fire compartment walls.
The disadvantages are:
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
•
•
•
•
The construction of walls, particularly in low- and medium-rise buildings, is slow
and less accurate than steelwork.
The walls are difficult to modify if alterations to the building are required in the
future.
They are a separate form of construction, which is likely to delay the contract
programme.
It is difficult to provide connections between steel and concrete to transfer the
large forces generated.
Recent developments in steel–concrete–steel composite sandwich construction
(Bi-steel®) largely eliminate these disadvantages and allow pre-fabricated and fully
assembled lift shafts to be erected simultaneously with the main steel framing.
Steel–concrete–steel construction can also be used for blast-resistant walls and
floors.
Braced-bay frames
Braced-bays are positioned in similar locations to reinforced concrete walls, so they
have minimal impact upon the planning of the building. They act as vertical trusses
which resist the wind loads by cantilever action.
The bracing members can be arranged in a variety of forms designed to carry
solely tension or alternatively tension and compression. When designed to take
tension only, the bracing is made up of crossed diagonals. Depending on the wind
direction, one diagonal will take all the tension while the other remains inactive.
Tensile bracing is smaller than the equivalent strut and is usually made up of flatplate, channel or angle sections. When designed to resist compression, the bracings
become struts and the most common arrangement is the ‘K’ brace.
The advantages of braced-bay frames are:
•
•
•
All beam-to-column connections are simple
The braced bays are concentrated in location on plan
The bracing configurations may be adjusted to suit planning requirements (eccentric bracing)
Steel Designers' Manual - 6th Edition (2003)
Anatomy of structure
•
•
•
•
The system is adjustable if building modifications are required in the future
Bracing can be arranged to accommodate doors and openings for services
Bracing members can be concealed in partition walls
They provide an efficient bracing system.
A disadvantage is:
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
•
Diagonal members with fire proofing can take up considerable space.
0
0
Fig. 2.31
Connections: (a) simple; (b) continuous; (c) semi-continuous
75
Steel Designers' Manual - 6th Edition (2003)
76
Multi-storey buildings
2.3.5 Connections
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
The most important aspect of structural steelwork for buildings is the design of the
connections between individual frame components.
The selection of a component should be governed not only by its capability to
support the applied load, but also by its ease of connection to other components.
Basically there are three types of connection, each defined by its structural behaviour: simple, continuous and semi-continuous (Fig. 2.31):
(1) Simple connections transmit negligible bending moment across the joint: the
connection is detailed to allow the beam end to rotate. The beam behaves as a
simply supported beam.
(2) Continuous connections are designed to transmit shear force and bending
moment across the joint. The connection should have sufficient stiffness or
moment capacity as appropriate to justify analysis by either elastic or plastic
analysis. Beam end moments are transmitted into the column itself and any
beam framing into the column on the opposite side.
(3) Semi-continuous connections are designed to transmit the shear force and a
proportion of the bending moment across the joint. The principle of these
connections is to provide a partial restraint to beam end-rotation without
introducing complicated fabrication to the joint. However, the design of such
joints is complex, and so simple design procedures based upon experimental
evidence have been developed for wider application. The advantages of semicontinuous design are lighter beams without the corresponding increase in
column size and joint complexity that would be the case with fully continuous
connections.
References to Chapter 2
1. Hart F., Henn W., Sontag H. & Godfrey G.B. (Ed.) (1985) Multi-Storey Buildings
in Steel, 2nd edn. Collins, London.
2. National Economic Development Office and Economic Development Committee for Constructional Steelwork (1985) Efficiency in the Construction of Steel
Framed Multi-Storey Buildings. NEDO, Sept.
3. Owens G. (1987) Trends and Developments in the Use of Structural Steel for
Multi-Storey Buildings. Steel Construction Institute, Ascot, Berks.
4. McGuire W. (1968) Steel Structures. Prentice-Hall.
5. Zunz G.J. & Glover M.J. (1986) Advances in Tall Buildings. Council on Tall Buildings and Urban Habitat. Van Nostrand Reinhold.
Steel Designers' Manual - 6th Edition (2003)
Further reading
77
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Further reading for Chapter 2
Brett P. & Rushton J. (1990) Parallel beam approach – a design guide. The Steel Construction Institute, Ascot, Berks.
Couchman G.H. (1997) Design of semi-continuous braced frames. The Steel Construction Institute, Ascot, Berks.
Hensman J.S. & Way A.G.J. (2000) Wind-moment design of unbraced composite
frames. The Steel Construction Institute, Ascot, Berks.
Lawson R.M. & Rackham J.W. (1989) Design of haunched composite beams in buildings. The Steel Construction Institute, Ascot, Berks.
Lawson R.M. & McConnel R. (1993) Design of stub girders. The Steel Construction
Institute, Ascot, Berks.
Lawson R.M., Mullett D.L. & Rackham J.W. (1997) Design of asymmetric Slimflor
beams using deep composite decking. The Steel Construction Institute, Ascot,
Berks.
Lawson et al. (2002) Design of FABSEC Beams in Non-Composite Applications
(Including Fire). The Steel Construction Institute, Ascot, Berks.
Mullett D.L. (1992) Slim floor design and construction. The Steel Construction Institute, Ascot, Berks.
Mullett D.L. (1998) Composite Floor Systems. Blackwell Science.
Mullett D.L. & Lawson R.M. (1999) Design of Slimflor fabricated beams using deep
composite decking. The Steel Construction Institute, Ascot, Berks.
Narayanan R., Roberts T.M. & Naji F.J. (1994). Design guide for steel–concrete–steel
sandwich construction – Volume 1: General principles and rules for basic elements.
The Steel Construction Institute, Ascot, Berks.
Owens G.W. (1989) Design of fabricated composite beams in buildings. The Steel
Construction Institute, Ascot, Berks.
Salter P.R., Couchman G.H. & Anderson D. (1999) Wind-moment design of low rise
frames. The Steel Construction Institute, Ascot, Berks.
Skidmore, Owings & Merrill (1992) Design of composite trusses. The Steel Construction Institute, Ascot, Berks.
Slimdek® Manual (2001) Corus Construction Centre, Scunthorpe, Lincs.
Ward J.K. (1990) Design of composite and non-composite cellular beams. The Steel
Construction Institute, Ascot, Berks.
Yandzio E. & Gough M. (1999) Protection of buildings against explosions. The Steel
Construction Institute, Ascot, Berks.
A worked example follows which is relevant to Chapter 2.
Steel Designers' Manual - 6th Edition (2003)
78
Worked example
Corus
Construction
Centre
www.corusconstruction.com
Subject
Chapter ref.
MULTI-STOREY DESIGN
EXAMPLE
Design code
BS 5950: Part 1: 2000
Made by
2
PEP
Checked by
Sheet no.
1
TRM
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Building geometry
Vertical
bracing
Typical
arrangement of
stair tower
Typical Floor Plan
Building use: Office building with basement car parking and high-level plant room.
Imposed loading for office floors exceeds minimum statutory loading given in BS
6399-1: 1996 at client’s request. This design example illustrates the design of elements
in the braced towers provided in four corners of the building to achieve lateral stability. The floor plate is generally 130 mm lightweight aggregate concrete on metal decking
that acts compositely with the decking and floor beams and is assumed to provide
diaphragm action. Fire protection is achieved with a sprayed intumescent coating. Alternatively, fire protection could be removed from a number of beams by adopting the fire
safe design approach outlined in Chapter 34.
Steel Designers' Manual - 6th Edition (2003)
Worked example
Subject
Corus
Construction
Centre
www.corusconstruction.com
Chapter ref.
MULTI-STOREY DESIGN
EXAMPLE
Design code
BS 5950: Part 1: 2000
Made by
2
PEP
Checked by
Sheet no.
2
TRM
kN/m2
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Roof loading
Dead
Steelwork + metal deck
Concrete (130 mm lw)
Finishes
Services below
0.27
1.80
2.00
0.30
4.37
Imposed
From BS 6399-3: 1988
(inc. Amd. 1–3 1997)
1.50
Plant room/B1 loading
Dead
Steelwork + metal deck
150 mm lw concrete slab
Suspended ceiling
Services
0.49
2.15
0.20 (not level B1)
0.30 (not level B1)
3.14 (2.64 – level B1)
Imposed
Floor
7.50 (2.50 – B1 level)
Dead
Steelwork + metal deck
130 mm lw concrete slab
Suspended ceiling
Raised floor
Services
0.27
1.80
0.20 (not ground)
0.30
0.30 (not ground)
2.87 (2.37 – ground)
Imposed
Floor
Partition allowance
4.00
1.00
Office loading
Stair loading
Dead
Imposed
As floor at level applied
4.00
79
Steel Designers' Manual - 6th Edition (2003)
Worked example
80
Subject
Corus
Construction
Centre
Chapter ref.
MULTI-STOREY DESIGN
EXAMPLE
www.corusconstruction.com
Made by
Design code
BS 5950: Part 1: 2000
2
PEP
Checked by
Sheet no.
3
TRM
Area supported
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Total floor area = 1688 m2
Total stair area = 158 m2
*Area supported (m2)
Col. G6
Col. F6
Col. F7
Floor
19.0
46.6
19.0
Stairs
0.8
1.6
0.8
* The above areas apply to all levels.
Perimeter loads
Dead
Roof
Plant room
General
Length
External
Atrium
Length supported by column
1.5 kN/m
18.8 kN/m
9.6 kN/m (not ground, B1 or B2)
120 + (3 ¥ 4) = 132.0 m
14.4 ¥ 4
= 57.6m
Total
189.6 m
1
2
3
=
=
=
4.8 m
0
4.8 m
(G6)
(F6)
(F7)
Notes:
The allocation of loads to columns 1, 2 and 3 based on the above data is shown on the
following sheets.
Imposed load reductions in accordance with BS 6399-1: 1966 are taken for all floors
but exclude the roof.
Steel Designers' Manual - 6th Edition (2003)
Worked example
Subject
Corus
Construction
Centre
Chapter ref.
MULTI-STOREY DESIGN
EXAMPLE
www.corusconstruction.com
81
Made by
Design code
PEP
Checked by
BS 5950: Part 1: 2000
2
Sheet no.
4
TRM
Column dead loads
Unit loads
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Level
R
Perim.
kN/m2
kN/m
kN
kN
kN
kN
kN
kN
4.37
1.5
93.7
210.6
(2)
93.7
93.7
210.6
93.7
3.14
18.8
(1)
152.4
151.3
152.4
246.1
(3)
361.9
246.1
2.87
9.6
102.9
138.3
102.9
349.0
500.2
349.0
2.87
9.6
102.9
138.3
102.9
451.9
638.5
451.9
2.87
9.6
102.9
138.3
102.9
554.8
776.8
554.8
2.87
9.6
102.9
138.3
102.9
657.7
915.1
657.7
2.87
9.6
102.9
138.3
102.9
760.6
1053.4
760.6
2.87
9.6
102.9
138.3
102.9
863.5
1191.7
863.5
2.87
9.6
102.9
138.3
102.9
966.4
1330.0
966.4
2.87
9.6
102.9
138.3
102.9
1069.3
1468.3 1069.3
2.37
46.9
114.2
46.9
1116.2
1582.5 1116.2
2.64
52.3
127.2
52.3
1168.5
1709.7 1168.5
8
≤8≤
≤7≤
≤6≤
≤5≤
≤4≤
≤3≤
≤2≤
≤1≤
4
4
4
4
4
4
4
4
Cumulative column loads
Floor
P
Column loads
Col.G6 Col.F6 Col.F7 Col.G6 Col.F6 Col.F7
6
G
B1
B2
4
4
Examples:
(1) (3.14 ¥ 19.0) + (3.14 ¥ 0.8) + (18.8 ¥ 4.8)= 152.4 kN
(2) (4.37 ¥ 46.6) + (4.37 ¥ 1.6)
= 210.6 kN
(3) 210.6 + 151.3
= 362.9 kN
Steel Designers' Manual - 6th Edition (2003)
Worked example
82
Subject
Corus
Construction
Centre
Chapter ref.
MULTI-STOREY DESIGN
EXAMPLE
www.corusconstruction.com
Made by
Design code
BS 5950: Part 1: 2000
2
PEP
Checked by
Sheet no.
5
TRM
Column imposed loads
Column loads
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Level
Imposed
load
kN/m2
R
Imposed Col.G6 Col.F6 Col.F7 Col.G6 Col.F6 Col.F7
load
reduction
%
kN
kN
kN
kN
kN
kN
1.5
0
31.7
76.3
31.7
31.7
76.3
31.7
7.5
0
(1)
145.7
(2)
355.9
145.7
177.4
432.2
177.4
5
10
98.2
239.4
98.2
251.2
612.1
251.2
5
20
98.2
239.4
98.2
305.4
305.4
5
30
98.2
239.4
98.2
339.9
744.1
(3)
828.2
5
40
98.2
239.4
98.2
354.8
864.4
354.8
5
40
98.2
239.4
98.2
413.7
1008.0 413.7
5
40
98.2
239.4
98.2
472.6
1151.7 472.6
5
40
98.2
239.4
98.2
531.6
1295.3 531.6
5
40
98.2
239.4
98.2
590.5
1439.0 590.5
5
40
98.2
239.4
98.2
649.4
1582.6 649.4
2.5
50
50.7
122.9
50.7
571.8
1393.0 571.8
8
P
≤8≤
≤7≤
≤6≤
≤5≤
≤4≤
≤3≤
≤2≤
≤1≤
4
4
4
4
4
4
4
4
*Cumulative column loads
339.9
6
G
B1
B2
4
4
* Values include % imposed load reductions
Examples:
(1) (7.5 ¥ 19.0) + (4.0 ¥ 0.8)
= 145.7 kN
(2) (7.5 ¥ 46.6) + (4.0 ¥ 1.6)
= 355.9 kN
(3) 76.3 + [(355.9 + 239.4 + 239.4 + 239.4) ¥ (1 - 30/100)] = 828.2 kN
Steel Designers' Manual - 6th Edition (2003)
Worked example
Corus
Construction
Centre
www.corusconstruction.com
Subject
Chapter ref.
MULTI-STOREY DESIGN
EXAMPLE
Design code
BS 5950: Part 1: 2000
83
2
PEP
Made by
Checked by
Sheet no.
6
TRM
Wind load
BS 6399-2: 1977 (inc. Amd. 1 2002)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Using standard method (since H < B division by parts is not applicable and frictional
drag is neglected as having little effect):
Basic wind speed Vb = 21 m/s (London)
Building length L = 48 m
Building width W = 48 m
Building wall height H = 46 m
Building reference height Hr = 46 m
Building type factor Kb = 1 (open plan office)
Dynamic augmentation factor Cr = 0.047
Altitude factor Sa = 1.02 (20 m above sea level)
Directional factor Sd = 1.00 (wind direction III – 210 deg)
Seasonal factor Ss = 1.00
Probability factor Sp = 1.00
øo,
.vII
1117
Site wind speed Vs = Vb Sa Sd Ss Sp = 21.42 m/s
Wind directions
Average height of nearby buildings Ho = 20 m
Upwind spacing Xo = 20 m
Displacement height Hd = 0.8Ho (Xo £ 2Ho)
Effective height He = greater of Hr - Hd or 0.4 Hr = 30.0 m
Distance to sea = 80 km (upwind @ 210 deg ± 45 deg)
Terrain = town
(Table 4 – by linear interpolation –
Terrain & building factor Sb = 1.874
Logarithmic interpolation is optional)
Effective wind speed Ve = Vs Sb = 40.15 m/s
Note: normally, either all wind directions should be checked to establish the highest
effective wind speed or a conservative approach may be taken by using a value of
Sd = 1.0 together with the shortest distance to sea irrespective of direction. A lower value
of Sb will be obtained for sites in town by using the hybrid approach.
Dynamic pressure qs = 0.613 V e2 = 988 N/m2
Breadth B = 48 m
Inwind depth D = 48 m
Ratio B/D = 1
Span ratio = D/H = 1.04 ≥ 1 but £ 4
2.4.1.2
Steel Designers' Manual - 6th Edition (2003)
84
Worked example
Subject
Corus
Construction
Centre
www.corusconstruction.com
Chapter ref.
MULTI-STOREY DESIGN
EXAMPLE
Made by
Design code
BS 5950: Part 1: 2000
Net pressure coefficient Cp = 1.195
2
PEP
Checked by
7
Sheet no.
TRM
(Table 5a – Linear interpolation)
Size effect factor for external pressure Ca = 0.839 Figure 4:
Effective height He = 30.0 m
Distance to sea = 80 km
Terrain = town
Dimension a = (482 + 46 2) / = 66.48 m
Site exposure type = A
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
1
2
For net wind load to building P = 0.85 ( SqsCpCa A)(1 + Cr ) Clause 2.1.3.6 NOTE 3
Simplifying
P = S881.6 A ¥ 10-3 kN
Loads will be applied to each storey and transferred to each floor level (see table
overleaf).
To provide a practical level of robustness against the effects of incidental loading, all
structures should have adequate resistance to horizontal forces. The horizontal component of the factored wind load should not be taken as less than 1.0% of the factored
dead load applied horizontally. (cl. 2.4.2.3)
Example to table on page 85
(1) (4.37 ¥ 1688) + (4.37 ¥ 158) + (1.5 ¥ 189.6) = 8351 kN
(2) 8351 ¥ 1.4 ¥ 1%
= 117 kN
Steel Designers' Manual - 6th Edition (2003)
Worked example
Subject
Corus
Construction
Centre
Chapter ref.
MULTI-STOREY DESIGN
EXAMPLE
www.corusconstruction.com
Made by
Design code
BS 5950: Part 1: 2000
Wind and minimum horizontal forces
h
Area Wind Force
P (kN)
(m) (m2) (Unfactored)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
R
F
46
384
338.5
192
169.3
192
169.3
192
169.3
192
169.3
192
169.3
192
169.3
192
169.3
192
169.3
288
253.9
85
2
PEP
Checked by
Sheet no.
8
TRM
Wind Force
F (kN)
(Unfactored)
Dead load
(kN)
(Unfactored)
Min. Force
F¢ (kN)
(Factored)
169.3
8351
(1)
117
(2)
253.9
9361
131
169.3
7118
100
169.3
7118
100
169.3
7118
100
169.3
7118
100
169.3
7118
100
169.3
7118
100
169.3
7118
100
211.6
7118
100
127.0
4375
61
4873
68
P
P
≤8≤
≤7≤
≤6≤
≤5≤
≤4≤
≤3≤
≤2≤
≤1≤
38
34
30
26
22
18
14
10
6
G
B1
B2
S1947
S1947
F is the total force applied at that level. The factored minimum force ex dead load (F¢)
is clearly less than the value of factored wind load and will be ignored in further
calculations.
Loads are divided by 4 in the sub-model analysis to represent the force applied to one
braced tower. To justify this approach the behaviour of the whole building was analysed
using CSC Fastrak® Multi-storey to confirm that the sub-model for the bracing system
was sufficiently accurate. One of the vertical braced towers was then analysed using CSC
S-FrameTM.
N.B. Asymmetric frames may be subject to torsional and out-of-plane P delta effects.
Steel Designers' Manual - 6th Edition (2003)
86
Worked example
Corus
Construction
Centre
Subject
Chapter ref.
MULTI-STOREY DESIGN
EXAMPLE
www.corusconstruction.com
Design code
BS 5950: Part 1: 2000
Made by
2
PEP
Checked by
Sheet no.
9
TRM
Notional horizontal forces (NHF)
To allow for the effects of practical imperfections such as lack of verticality, notional
horizontal forces are considered.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
At each level,
FNHF = 0.5% of the factored vertical dead and imposed loads at that level.
For simplicity, calculations for the NHFs do not include any % imposed load
reductions.
Roof
D.L.
L.L.
NHF
= 4.37 ¥ (1688 + 158) + (1.5 ¥ 189.6)
=
= (1.5 ¥ 1688) + (4.0 ¥ 158)
=
= 0.5/(100 ¥ 4) ¥ [(1.4 ¥ 8351) + (1.6 ¥ 3164)] =
kN
8351
3164
20.9
Plant room
D.L.
L.L.
NHF
= 3.14 ¥ (1688 + 158) + (18.8 ¥ 189.6)
=
= (7.5 ¥ 1688) + (4.0 ¥ 158)
=
= 0.5/(100 ¥ 4) ¥ [(1.4 ¥ 9361) + (1.6 ¥ 13292)] =
9361
13292
43.0
Office floors (typ.)
D.L.
L.L.
NHF
= 2.87 ¥ (1688 + 158) + (9.6 ¥ 189.6)
=
= (5 ¥ 1688) + (4.0 ¥ 158)
=
= 0.5/(100 ¥ 4) ¥ [(1.4 ¥ 7118) + (1.6 ¥ 9072)] =
7118
9072
30.6
Ground
D.L.
L.L.
NHF
= 2.37 ¥ (1688 + 158)
=
= (5 ¥ 1688) + (4.0 ¥ 158)
=
= 0.5/(100 ¥ 4) ¥ [(1.4 ¥ 4375) + (1.6 ¥ 9072)] =
4375
9072
25.8
Basement 1
D.L.
L.L.
NHF
= 2.64 ¥ (1688 + 158)
=
= (2.5 ¥ 1688) + (4.0 ¥ 158)
=
= 0.5/(100 ¥ 4) ¥ [(1.4 ¥ 4873) + (1.6 ¥ 4852)] =
4873
4852
18.2
Steel Designers' Manual - 6th Edition (2003)
Worked example
Subject
Corus
Construction
Centre
Chapter ref.
MULTI-STOREY DESIGN
EXAMPLE
www.corusconstruction.com
Design code
BS 5950: Part 1: 2000
Made by
2
PEP
Checked by
Sheet no.
10
TRM
Summary of NHF loads
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Level
Roof
Plant
8
7
6
5
4
3
2
1
Ground
Base 1
Base 2
NHF
kN
20.9
43.0
30.6
30.6
30.6
30.6
30.6
30.6
30.6
30.6
25.8
18.2
The NHF load case is considered as acting simultaneously with
dead and imposed loads when no wind forces are acting.
N.B. The NHFs should be applied in the two orthogonal directions (global X and Y) separately and may need to be applied
in the positive and negative direction to give the worst effect
Sway stiffness P–d effects
All structures (including portions between expansion joints) should
have sufficient sway stiffness, so that the vertical loads acting with
the lateral displacements of the structure do not induce excessive
secondary forces or moments in the members or connections. Where
such ‘second order’ (P–d) effects are significant, they should be
allowed for in the design of those parts of the structure that contribute to its resistance to horizontal forces.
2.4.2.5
Sway stiffness should be provided by the system of resisting horizontal forces. Whatever system is used, sufficient stiffness should be
provided to limit sway deformation in any horizontal direction and
also to limit twisting of the structure on plan.
Except for single-storey frames with moment-resisting joints, or
other frames in which sloping members have moment-resisting
connections, lcr should be taken as the smallest value, considering
every storey.
In the following table, the elastic critical buckling load factor lcr
has been determined from:
lcr = h / ( 200 ¥ d )
where
h
d
= storey height and
= storey drift due to NHFs
2.4.2.6
87
Steel Designers' Manual - 6th Edition (2003)
Worked example
88
Subject
Corus
Construction
Centre
Chapter ref.
MULTI-STOREY DESIGN
EXAMPLE
www.corusconstruction.com
Design code
BS 5950: Part 1: 2000
Made by
2
PEP
Checked by
Sheet no.
11
TRM
Because of symmetry one of the four braced towers only was modelled as a space frame
analysed using CSC S-FrameTM.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Basic load cases
1.
2.
3.
4.
Dead load
Reduced imposed loads
Wind loading (Y direction)
NHF (Y direction)
Y
Results from NHF load case
Lateral
disp.
mm
Level
R
≤8≤
≤7≤
≤6≤
≤5≤
≤4≤
≤3≤
≤2≤
≤1≤
4
4
4
4
4
4
4
4
-39.02
-35.72
-32.26
-28.66
-24.99
-21.33
-17.67
-14.05
-10.54
-5.70
-2.70
0
6
G
B1
B2
4
4
G6
Storey drift
d
mm
Line of application
of wind loads
lcr
PLAN
Line of application
of vertical loads
-45.02
8
P
F7
F6
6.00
6.67
3.30
6.06
3.46
5.78
3.60
5.56
3.67
5.45
3.66
5.46
3.66
5.46
3.62
5.52
3.51
5.70
4.84
6.20
All nodes at G6 &
F6 are restrained in
x direction to model
restraints provided
by main building
= restraint
1r
3.00
6.67
2.70
7.41
_______
F6
•1
/F7
DIAGRAM OF
TOWER MODEL
Steel Designers' Manual - 6th Edition (2003)
Worked example
Corus
Construction
Centre
www.corusconstruction.com
Subject
Chapter ref.
MULTI-STOREY DESIGN
EXAMPLE
Design code
BS 5950: Part 1: 2000
Made by
2
PEP
Checked by
The smallest value of lcr is 5.45, which is < 10. Therefore the frame
is sway sensitive, and as lcr is not less than 4.0, the amplified sway
method may be used.
This gives an amplification factor for NHFs in the Y direction of:
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Sheet no.
12
TRM
kamp = lcr / (1.15 lcr - 1.5 ) = 1.14
(for clad structures, provided that the stiffening effect of masonry
infill wall panels or diaphragms of profiled steel sheeting is not
explicitly taken into account.)
Cautionary notes to analysis approach
1. Results are for sway only in the Y direction using NHF in the
Y direction from the gravity combination. This analysis must
normally be repeated for sway in the X direction using NHF in
the X direction. The choice is then to either:
(a) use the smallest value of lcr to determine one value of kamp
applied to loads in both directions, or
(b) calculate a lcr and hence kamp for each direction.
Similarly a kamp could be calculated for each load combination
and applied to each appropriate combination.
2. The results from a simple tower constrained out-of-plane at all
levels take no account of out-of-plane deflections and their
influence on P–d effects. An analysis of the whole building is
required to spot this effect and determine whether it is significant. In this example the tower constrained out of plane is only
a reasonable assumption owing to the very symmetric nature of
the building as a whole. If the whole building is considered then
there could be other columns (even simple ones not in braced
bays) that lean over more. These will have a detrimental effect
on the braced areas. That is, whilst not affected themselves
(being simple columns), they will increase the P–d effect on the
braced towers. Taking the worst sway index for all columns also
helps to compensate for any out-of-plane deflections.
2.4.2.7
89
Steel Designers' Manual - 6th Edition (2003)
Worked example
90
Corus
Construction
Centre
www.corusconstruction.com
Subject
Chapter ref.
MULTI-STOREY DESIGN
EXAMPLE
Made by
Design code
BS 5950: Part 1: 2000
2
PEP
Checked by
Sheet no.
13
TRM
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
3. The horizontal members of the tower frame are constrained by
the floor diaphragm in the whole building model. There are two
effects.
(a) The tower model has to be given horizontal members that
are significantly stiffer than the beam stiffness alone. In the
model analysed the 406 ¥ 178 ¥ 74 UB was retained but
given an E value of 1000 times greater than that for steel.
(b) In the whole building model the diaphragm transmits the
horizontal loads to the tower and so there is no axial load
present in the horizontal beams.
2.4.2.4
Load case combinations
The notional horizontal forces should be taken as acting simultaneously with the factored vertical dead and imposed loads but
should not be combined with applied horizontal loads.
The load combinations are therefore:
1.4 ¥ Dead + 1.6 ¥ Imposed
1.2 ¥ ( Dead + Imposed ± kamp ¥ Wind )
1.4 ¥ Dead ± 1.4 ¥ kamp ¥ Wind
1.0 ¥ Dead ± 1.4 ¥ kamp ¥ Wind
1.4 ¥ Dead + 1.6 ¥ Imposed ± kamp ¥ NHF (Y -direction)
Proposed section sizes
The columns have been spliced at several locations to ensure practical changes in section size and allow transport. The sections have
been chosen to have sufficient stiffness to give reasonable sway
behaviour. This means that the sections are over-designed for
strength and buckling – see Sheet 15.
Stacks
Stacks
Stacks
Stacks
1–3, B2 – Floor 1
4–6, Floor 1 – Floor 4
7–9, Floor 4 – Floor 7
10–12, Floor 7 – Roof
356
356
356
305
¥
¥
¥
¥
406
406
368
305
¥
¥
¥
¥
340 UC
235 UC
129 UC
97 UC
Braces are HF 250 ¥ 250 ¥ 8 SHS and floor beams are 406 ¥ 178
¥ 74 UB (see cautionary note 3(a) above).
Deflections under wind loading are given in the table below.
Steel Designers' Manual - 6th Edition (2003)
Worked example
Subject
Corus
Construction
Centre
Chapter ref.
MULTI-STOREY DESIGN
EXAMPLE
www.corusconstruction.com
91
Design code
BS 5950: Part 1: 2000
Made by
2
PEP
Checked by
Sheet no.
14
TRM
Lateral deflections due to wind load
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Level
R
Lateral
disp.
Storey drift
d
(mm)
(mm)
≤8≤
≤7≤
≤6≤
≤5≤
≤4≤
≤3≤
≤2≤
≤1≤
4
4
4
4
4
4
4
4
-58.1
-53.0
-47.7
-42.2
-36.7
-31.2
-25.8
-20.4
-15.3
-8.1
-3.8
0.0
6
G
B1
B2
4
4
d
-67.5
8
P
Height
9.4
851
5.1
784
5.3
755
5.5
727
5.5
727
5.5
727
5.4
741
5.4
741
5.1
784
7.2
833
4.3
930
3.8
1053
G6
Maximum differential storey deflection due to wind loads is h/727 < h/300
Therefore, deflection is acceptable.
F7
Steel Designers' Manual - 6th Edition (2003)
92
Worked example
Subject
Corus
Construction
Centre
Chapter ref.
MULTI-STOREY DESIGN
EXAMPLE
www.corusconstruction.com
Design code
BS 5950: Part 1: 2000
Made by
2
PEP
Checked by
Sheet no.
15
TRM
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Confirmation of member sizes
The columns can be treated as simple even though they are part of braced bays. The
moments due to continuity of the columns are small and are ignored. The eccentricity
moments due to the end reactions from the incoming beams are considered. As simple
columns they are designed to Clause 4.7.7 of BS 5950: Part 1: 2000. The beams are
designed as simply-supported with no axial load and the braces are designed for axial
load only.
The results for the most critical members in the braced tower are given below.
Member
Section size
Critical
position
Critical combination
Utilization
ratio
Column F6
356 406 340 UC
Below Floor 1
1.4 D + 1.6 I + NHF Y
0.65
Column F6
356 406 235 UC
Above Floor 1
1.4 D + 1.6 I + NHF Y
0.67
Column F6
356 368 129 UC
Below Floor 5
1.4 D + 1.6 I + NHF Y
0.81
Column F6
305 305 97 UC
Below Floor 8
1.4 D + 1.6 I + NHF Y
0.67
Office floor beam
406 178 74 UB
Mid-span
1.4 D + 1.6 I
0.85
Plant floor beam
533 210 82 UB
Mid-span
1.4 D + 1.6 I
0.90
Brace – floor 1 to G HF250 250 8 SHS
Compression
1.4 kamp Wind
0.74
Brace – floor 1 to G HF250 250 8 SHS
Tension
1.4 kamp Wind
0.44
All sections in S275. SHS are hot-finished to BS EN 10210.
Steel Designers' Manual - 6th Edition (2003)
Worked example
Subject
Corus
Construction
Centre
www.corusconstruction.com
Chapter ref.
MULTI-STOREY DESIGN
EXAMPLE
Design code
BS 5950: Part 1: 2000
Made by
2
PEP
Checked by
Sheet no.
16
TRM
Robustness of frame
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
To provide a degree of robustness and to reduce the possibility of
accidental damage, column ties should be capable of transmitting
(1) The general tying force given by 0.5 wf st La for internal ties or
0.25 wf st La for edge ties but not less than 75 kN, where:
wf
=
factored dead and imposed load per unit area
sf
=
mean transverse spacing of ties
La
=
maximum span within the overall length of the tie
tying force (office ) = 0.5 ¥ [( 2.87 ¥ 1.4 ) + (5.0 ¥ 1.6 )] ¥ 7.5 ¥ 7.5
= 338 kN > 75 kN
or
(2) At the periphery, the general tying force or an anchor force of
1% of the factored vertical load in the column, whichever is
the greater.
Max . factored column force (Cols. Cl , C7 , E 1 and C7 ) = 6360 kN
Therefore anchor force = 63.60 kN
Design connection for P = ± 338 kN
(NB these forces should not be considered as additive to other loads.)
:
The author wishes to thank Mr Alan Rathbone of CSC (UK) Ltd.
for his assistance in the preparation of this worked example.
2.4.5.3
93
Steel Designers' Manual - 6th Edition (2003)
Chapter 3
Industrial steelwork
by JOHN ROBERTS
3.1 Range of structures and scale of construction
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
3.1.1 Introduction
Structural steelwork for industrial use is characterized by its function, which is primarily concerned with the support, protection and operation of plant and equipment. In scale it ranges from simple support frameworks for single tanks, motors or
similar equipment, to some of the largest integrated steel structures, for example
complete electric power-generating facilities.
Whereas conventional single- and multi-storey structures provide environmental
protection to space enclosed by walls and roof and, for multi-storey buildings,
support of suspended floor areas, these features are never dominant in industrial
steelwork. Naturally, in many industrial structures, the steel framework also
provides support for wall and roof construction to give weather protection, but
where this does occur the wall and roof profiles are designed to fit around and suit
the industrial plant and equipment, frequently providing lower or different standards of protection in comparison with conventional structures. Many plant installations are provided only with rain shielding; high levels of insulation are unusual,
and some plant and equipment is able to function and operate effectively without
any weather protection at all. In such circumstances the requirements for operational and maintenance personnel dictate the provision of cladding, sheeting or
decking.
Similarly, most industrial steelwork structures have some areas of conventional
floor construction, but this is not a primary requirement and the flooring is incidental to the plant and equipment installation.
Floors are provided to allow access to and around the installation, being arranged
to suit particular operational features. They are therefore unlikely to be constructed
at constant vertical spacing or to be laid out on plan in any regular repetitive pattern.
Steelwork designers must be particularly careful not to neglect the importance of
two factors. First, floors cannot automatically be assumed to provide a horizontal
wind girder or diaphragm to distribute lateral loadings to vertical-braced or framed
bays: openings, missing sections or changes in levels can each destroy this action.
Secondly, column design is similarly hampered by the lack of frequent and closely
spaced two-directional lateral support commonly available in normal multi-storey
structures.
94
Steel Designers' Manual - 6th Edition (2003)
Range of structures and scale of construction
95
Floors require further consideration regarding the choice of construction (see
section 3.2.3) and loading requirements (see section 3.3.1).
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
3.1.2 Power station structures
Industrial steelwork for electrical generating plants varies considerably depending
on the size of station and the fuel being used. These variations are most marked in
boiler house structures; whereas coal-fired and oil-fired boilers are similar, nuclear
power station boilers (reactors) are generally constructed in concrete for biological
shielding purposes, steel being used normally in a secondary building envelope role.
Turbine halls are, in principle, largely independent of fuel type, and many of the
other plant structures (mechanical annexes, electrical switchgear buildings, coal
hoppers, conveyors, pump-houses) are common in style to other industrial uses and
so brief descriptions of the salient design features are of general interest.
Boiler houses (coal- or oil-fired) (Fig. 3.1) have to solve one overriding design
criterion and as a result can be considered exercises in pure structural design.
Fig. 3.1
Plan of boiler house framing
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
96
Industrial steelwork
Modern boilers are huge single pieces of plant with typical dimensions of 20 m ¥
20 m ¥ 60 m high for a single 500 MW coal boiler. Where poor-quality coal is burnt
or higher capacities are required, the dimensions can be even larger, up to about
25 m ¥ 25 m ¥ 80 m high for 900 MW size sets. As may be anticipated, the weights
are equally massive, typically in the range of 7000–10 000 t for the plant sizes noted
above.
Boilers are always top-suspended from the supporting structures and not built
directly from foundation level upwards, nor carried by a combination of top and
bottom support. This is because the thermal expansion of the boiler prevents dual
support systems; unsurmountable buckling and stability problems on the thinwalled-tube structure of the casing would arise if the boiler was bottom-supported
and hence in compression, rather than top-suspended and hence in tension. For
obvious reasons no penetrations of the boiler can be acceptable and therefore no
internal columns can be provided. The usual structural system is to provide an
extremely deep and stiff system of suspension girders (plate or box) spanning across
the boiler with an extensive framework of primary, secondary and (twin) tertiary
beams terminating in individual suspension rods or hangers which support the
perimeter walls and roof of the boiler itself.
Columns are massively loaded from the highest level and so are usually constructed from welded box-sections since their loading will be considerably above
the capacity of rolled sections. It is usual practice for a perimeter strip some 5–10 m
wide to be built around the boiler itself allowing a structural grid to be provided with
an adequate bay width for bracing to be installed for lateral stability. It also provides
support for ancillary plant and equipment adjacent to specific zones of the boiler,
for pipework and valves, for personnel access walkways or floor zones.
The pipework support requirements are often onerous and in particular the
pipework designers may require restrictive deflection limitations that are sometimes
set as low as 50 mm maximum deflection under wind loadings at the top of 90 m
high structures.
Turbine halls (Fig. 3.2) support and house turbo-generating machines that operate
on steam produced by the boiler, converting heat energy into mechanical energy of
rotation and then into electrical energy by electromagnetic induction. Turbogenerators are linear in layout, built around a single rotating shaft, typically some
25 m long for 500 MW units. The function of a steel-framed turbine hall is to protect
and allow access to the generator, to support steam supply and condensed water
return pipework and numerous other items of ancillary plant and equipment. Heavy
crane capacity is usually provided since generators are working machines that
require routine servicing as well as major overhaul and repairs. They are probably
the largest rotating machines in common use, and dynamic analysis of turbine generator support steelwork is imperative. Fortunately they operate at a sensibly constant rotational velocity and so design of the support steelwork is amenable to an
analytical examination of dynamic frequencies of motion of the whole support structure with plant loading in each possible mode, with similar consideration of any local
frequency effect, such as vibration of individual elements of the structure or its
framework.
Steel Designers' Manual - 6th Edition (2003)
Range of structures and scale of construction
97
—
LJ
overhead travelling crane
crone gkder
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
IonçtudinaI
bracing
Fig. 3.2
Cross section through turbine hall
These frequencies of response are compared with the forcing frequency of motion
of the machine, the design being adjusted so that no response frequencies exist in
a band either side of the forcing frequency, to avoid resonance effects.
A complete physical separation is provided between the generator support steelwork and any adjoining main frame, secondary floor or support steel to isolate vibration effects. Zones on adjoining suspended floors are set aside for strip-down and
servicing, and specific ‘laydown’ loading rates are allowed in the design of these
areas to cater for heavy point and distributed loads.
3.1.3 Process plant steelwork
Steelwork for process and manufacturing plants varies across a wide spectrum of
different industrial uses. Here it is considered to be steelwork that is intimately con-
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
98
Industrial steelwork
nected with the support and operation of plant and equipment, rather than simply
a steel-framed building envelope constructed over a process plant.
Although the processes and plant vary widely, the essential features of this type
of industrial steelwork are common to many applications and are conveniently
examined by reference to some typical specific examples. There are many similarities with power station boiler house and turbine house steelwork described in
section 3.1.2.
Cement manufacturing plant. Typical cement plants are an assembly of functional
structures arranged in a manufacturing flow sequence, with many short-term storage
and material transfer facilities incorporated into the processes. The physical height
and location of the main drums are likely to dictate the remaining plant
orientation.
Vehicle assembly plants. Substantial overhead services to the various assembly
lines characterize vehicle assembly plants. It is normal to incorporate a heavy-duty
and closely-spaced grid roof structure which also will support the roof covering.
Reasonably large spans are needed to allow flexibility in arranging assembly line
layouts without being constrained by column locations. Automation of the assembly process brings with it stiffness requirements to allow use of robots for precise
operations such as welding and bonding. Open trusses in two directions are likely
to satisfy most of these requirements, providing structural depth for deflection
control and a zone above bottom boom level that can be used for service runs. Building plans are normally regular with rectangular type plan forms and uniformly
regular roof profiles.
Nuclear fuel process and treatment plants. Steelwork for nuclear fuel process and
treatment plants is highly dependent on the actual process involved, and often has
to incorporate massive concrete sections for biological radiation shielding purposes.
Particular points to note are the importance of the paint or other finishes, both from
the point of view of restricted access in certain locations, leading to maintenance
problems, and also from the necessity for finishes in some areas to be capable of
being decontaminated. Specialized advice is needed for the selection of suitable finishes or to give guidance on whether, for example, structural stainless steels would
be appropriate. In addition, certain nuclear facilities must be designed for extreme
events, the most relevant of which is seismic loading set by the statutory regulatory
body. Frequently designs will have to be undertaken to comply with well-established
codes of practice for seismically active zones in, for example, America. Seismic
design requires the establishment of ductile structures to allow high levels of energy
absorption prior to collapse, and local stability and ductile connection behaviour
become of critical importance. It follows that joint and connection design has to be
fully integrated with the structural steel design generally; the normal responsibilities of joint design assumed by the steelwork fabricator may have to be altered on
these projects.
Petrochemical plants. Petrochemical plants tend to be open structures with little
or no weather protection. The steelwork required is dedicated to providing support
to plant and pipework, and support for access walkways, gangways, stairs and
ladders. Plant layouts are relatively static over long periods of use, and the steel-
Steel Designers' Manual - 6th Edition (2003)
Range of structures and scale of construction
99
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
work is relatively economic in relation to the equipment costs. It is normal therefore to design the steelwork in a layout exactly suited to the plant and equipment
without regard for a uniform structural grid. Benefits can still be gained from standardization of sizes or members and from maintaining an orthogonal grid to avoid
connection problems. The access floors and interconnecting walkways and stairways
must be carefully designed for all-weather access; use of grid flooring is almost universal. Protection systems for the steelwork should acknowledge both the threat
from the potentially corrosive local environment due to liquid or gaseous emissions
and the requirements to prevent closing down the facility for routine maintenance
of the selected protection system.
Careful account needs to be taken of wind loading in terms of the loads that occur
on an open structure and the lack of well-defined horizontal diaphragms from conventional floors. These two points are discussed further in sections 3.3.4 and 3.2.2.
3.1.4 Conveyors, handling and stacking plants
Many industrial processes need bulk or continuous handling of materials with a
typical sequence as follows:
(1) loading from bulk delivery or direct from mining or quarrying work
(2) transportation from bulk loading area to short-term storage (stacking or
holding areas)
(3) reclaim from short-term storage and transport to process plant.
Structural steelwork for the industrial plant which is utilized in these operations is
effectively part of a piece of working machinery (Fig. 3.3). Design and construction
standards must recognize the dynamic nature of the loadings and particularly must
cater for out-of-balance running, overload conditions and plant fault or machinery
failure conditions, any of which can cause stresses and deflections significantly
higher than those resulting from normal operation. Most designers would adopt
slightly lower factors of safety on loading for these conditions, but decisions need
to be based on engineering judgement as to the relative frequency and duration of
these types of loadings. The plant designer may well be unaware that such design
decisions can be made for the steelwork and frequently will provide single
maximum loading parameters that could incorporate a combination of all such possible events rather than a separate tabulation; the structural steelwork designer who
takes the trouble to understand the operation of the plant can therefore ask for the
appropriate information and use it to the best advantage.
A further feature of this type of steelwork is the requirement for frequent
relocation of the loading area and hence the transportation equipment. Practical
experience suggests that precise advance planning for specific future relocation is
rarely feasible, so attention should be directed to both design and detailing so that
future moves can cause the least disruption.
Steel Designers' Manual - 6th Edition (2003)
100
Industrial steelwork
conveyor
late rd
stability
bracing
braced conveyor
support
/
___________
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
support
bearing
/N
support
frame
elevation
section
A—A
LA
Fig. 3.3
//
_____________________
Typical details of conveyor support
Foundation levels can be set at constant heights; or, if variations have to occur,
then modular steps above or below a standard height should be adopted. The route
should utilize standard plan angles between straight sections, and uniform vertical
sloping sections between horizontal runs. Common base plate details and foundation bolt details can be utilized, even where this may be uneconomic for the initial
layout installation. Some consideration should be given to allowing the supporting
structure to be broken down into conveniently handled sections rather than into
individual elements; the break-down joints can be permanently identified, for
example, by painting in different colours.
3.1.5 General design requirements
Fatigue loadings must be considered, but even where fatigue turns out not to be a
design criterion, it is vital to protect against vibration and consequent loosening of
bolted assemblies. In this respect HSFG bolts are commonly used as they dispense
with the necessity to provide a lock-nut; when correctly specified and installed they
also display good fatigue-resistance.
Steel Designers' Manual - 6th Edition (2003)
Anatomy of structure
101
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
However, just as plant engineers can display a lack of understanding about the
necessity to design supporting steelwork afresh for each different structure even
though the ‘same’ plant is being built, structural steelwork engineers must be aware
that seemingly identical or repeat pieces of plant or equipment can in fact vary in
significant details. Most plant installations are purpose-designed, and the layout and
loadings are often provided initially to the structural engineer in terms of estimated
or approximate values. It is vital to be aware of the accuracy of the information
being used for design at any stage, and to avoid carrying out designs at an inappropriately advanced level. A considerable margin should be allowed, provided that
it is established that the plant designers have not allowed a similar margin already
in estimating the plant loadings. Experience shows that loadings are often overestimated at the preliminary design stage, but that this is compensated for by new
loadings at new locations that were not originally envisaged.
3.2 Anatomy of structure
3.2.1 Gravity load paths
Vertical loadings on industrial steelwork can be extraordinarily heavy; some individual pieces of plant have a mass of 10 000 t or more. Furthermore, by their very
nature these loadings generally act as discrete point loads or line loads rather than
as uniformly distributed loadings. Load values are often ill-defined (see section
3.3.1) at the steelwork design stage and frequently additional vertical loadings are
introduced at new locations late in the design process.
For these reasons, the gravity load paths must be established at an early stage to
provide a simple, logical and well-defined system. The facility should exist to cater
for a new load location within the general area of the equipment without the need
to alter all existing main structural element locations. Typically this means that it is
best to provide a layered system of beams or trusses with known primary span directions and spacings, and then with secondary (and sometimes in complex layouts,
tertiary as well) beams, which actually provide vertical support to the plant.
While simplicity and a uniform layout of structure are always attractive to a structural designer, the non-uniform loadings and layout of plant mean that supporting
columns may have to be positioned in other than a completely regular grid to
provide the most direct and effective load path to the foundations. It is certainly
preferable to compromise on a layout that gives short spans and a direct, simple
route of gravity load to columns, than to proceed with designing on a regular grid
of columns only to end up with a large range of member sizes and even types of
beams, girders or trusses (Fig. 3.4).
Similarly, although it is clearly preferable for columns to run consistently down
to foundation level, interference at low levels by further plant or equipment is quite
common, which may make it preferable to transfer vertical loading to an offset
column rather than use much larger spans at all higher levels in the structure. A
Steel Designers' Manual - 6th Edition (2003)
102
Industrial steelwork
column layout specific to
plant item, ignoring uniform
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
___1—
LH•.
iteT
...
plant
shown shaded
Fig. 3.4
Support steelwork for an unsymmetrical plant item
conscious effort to be familiar with all aspects of the industrial process, and close
liaison with the plant designers so that they are aware of the importance of an early
and inviolate scheme for column locations, are both necessary to overcome this
problem.
Different plant designers may be handling the equipment layout at differing levels
in the structure, so it is important to ensure that loading information from all parties
is received before the crucial decision on column spacing is made.
If a widely variable layout of potential loading attachment points exists, then
consideration must be given to hanging or top-supporting certain types of plant.
A typical example of this is a vehicle assembly line structure, where the overhead
assembly line system of conveyors can sensibly be supported on a deep truss roof
Steel Designers' Manual - 6th Edition (2003)
Anatomy of structure
103
with a regular column system. If this is contrasted with the multitude of columns
and foundations that would be needed to support such plant from below and the
prospect of having to reposition these supports during the design stage as more
accurate information on the plant becomes available, then it can be an appropriate,
if unorthodox, method of establishing a gravity load path.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
3.2.2 Sway load paths
For two fundamental reasons sway load paths require particular consideration in
industrial steelwork. First, the plant or equipment itself can produce lateral loading
on the structure in addition to wind or seismic load where applicable. Therefore,
higher total lateral loadings can exist whose points of application may differ from
the usual cladding and floor intersection locations. The type and magnitudes of such
additional loads are covered in section 3.3.3.
Second, many industrial steelwork structures lack a regular and complete (in
plan) floor construction that provides a convenient and effective horizontal
diaphragm. Therefore, the lateral load transfer mechanism must be considered carefully at a very early stage in the design (Fig. 3.5).
plant item
(no floor)
diaphragm action
*
Fig. 3.5
diaphragm action
vertical bracing
lateral stability provided at other levels
Establishment of sway load paths
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
104
Industrial steelwork
Naturally, there are many different methods of achieving lateral stability. Where
floor construction is reasonably complete and regular through the height of the
structure, then the design can be based on horizontal diaphragms transferring load
to vertical stiff elements at intervals along the structure length or width. Where large
openings or penetrations exist in otherwise conventional concrete floors, then it is
important to design both the floor itself and the connections between the floor and
steelwork for the actual forces acting, rather than simply relying automatically on
the provision of an effective diaphragm as would often and justifiably occur in the
absence of such openings.
It can be worthwhile deliberately to influence the layout to allow for at least a
–1
reasonable width of floor along each external face of building, say of the order of 10
of the horizontal spacing between braced bays or other vertical stiff elements.
When concrete floors as described above are wholly or sensibly absent, then other
types of horizontal girders or diaphragms can be developed. If solid plate or open
mesh steel flooring is used, then it is possible, in theory at least, to design such flooring as a horizontal diaphragm, usually by incorporating steel beams as ‘flange’
members of an idealized girder, where the floor steel acts as the web. However, in
practice, this is usually inadvisable as the flooring plate fixings are rarely found to
be adequate for load transfer and indeed it may be a necessary criterion that some
or all of the plates can be removable for operational purposes. A further factor is
that any line of beams used as a ‘flange’ must be checked for additional direct
compression or tension loadings; the end connections also have to be designed to
transfer these axial forces.
It is therefore normal to provide plan bracing in steelwork in the absence of concrete floor construction. The influence that this will have on plant penetrations,
pipework routeing and many other factors must be considered at an early stage in
liaison with the plant engineers. Naturally, the design considerations of steel beams
serving a dual function as plan bracing are exactly as set out above and must not
be neglected. Indeed, it may be preferable to separate totally the lateral load
restraint steelwork provided on plan from other steelwork in the horizontal plane.
This will avoid such clashes of purposes and clearly signal to the plant designers the
function of the steelwork, as a result preventing its misuse or abuse at a later stage.
Many examples exist where plan bracing members have been removed owing to
subsequent plant modifications. A practical suggestion to further minimize the possibility of this happening is to paint such steelwork a completely different colour as
well as to separate it completely from any duality of function with respect to plant
support or restraint.
Where it proves impossible to provide any type of horizontal plan bracing, then
each and every frame can be vertically braced or rigid-framed down to foundations.
It is best not to mix these two systems if possible, since they have markedly differing stiffnesses and will thus deflect differently under loading.
If diaphragms or horizontal girders are used, then the vertical braced bays that
receive lateral loading as reactions from them are usually braced in steelwork. In
conventional structures tension-only ‘X’ bracing is frequently used, and whereas this
may be satisfactory for some straightforward industrial structures such as tank
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Anatomy of structure
105
support frames and conveyor support legs, it often proves necessary to design combined tension/compression bracing in ‘N’, ‘K’, ‘M’ or similar layouts depending on
the relative geometry of the height to width of the bay and on what obstruction the
bracing members cause to plant penetrations. Indeed, it is sound advice initially to
provide significantly more bracing than may be considered necessary, for example
by bracing in two, three or more bays on one line. When later developments in the
plant and equipment layout mean that perhaps one or more panels must be altered
or even removed, it is then still possible to provide lateral stability by rechecking
or redesigning the bracing, without a major change in bracing location.
One particular aspect of vertical bracing design that requires care is in the evaluation of uplift forces in the tension legs of braced frames. Where plant and equipment provide a significant proportion of the total dead load, then it is important
that minimum dead weights of the plant items are used in calculations. For virtually
every other design requirement it is likely that rounding-up or contingency additions to loadings will have been made, especially at the early stages of the design.
It is also important to establish whether part of the plant loading is a variable contents weight and, if so, to deduct this when examining stability of braced frames. In
some cases, for example in hoppers, silos or tanks, this is obvious; but boilers or turbines which normally operate on steam may have weights expressed in a hydraulic
test condition when flooded with water. The structural engineer has to be aware of
these plant design features in order to seek out the correct data for design (see
section 3.3.2).
3.2.3 Floors
As has already been discussed in section 3.2.2, regular and continuous concrete
floors are not usual in industrial steelwork structures. Floors can consist of any of
the following types:
(1)
(2)
(3)
(4)
(5)
(6)
(7)
In situ concrete (cast on to removable formwork)
In situ concrete (cast on to metal deck formwork)
Fully precast flooring (no topping)
Precast concrete units with in situ concrete topping
Raised pattern ‘Durbar’ solid plate
Flat solid steel plate
Open grid steel flooring.
Typical sections of these types of floor construction are shown in Fig. 3.6.
The selection of floor type depends on the functional requirements and anticipated usage of the floor areas. Solid steel plates are used where transit or infrequent
access is required or where the floor must be removable for future access to plant
or equipment. They are normally used only internally (at least in the UK) to avoid
problems with wet or waterlogged surfaces.
Steel Designers' Manual - 6th Edition (2003)
106
Industrial steelwork
(1)
(2)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(4)
(5,6)
Fig. 3.6
________
IU
ii
IIIIuI
(7)
!U
Typical sections of floor construction
Open grid flooring is used inside for similar functions as solid steel plate, but with
additional functional requirements where the flooring is subjected to spillage of
liquids or where air flow through the floor is important. It is also used for stair treads
and landings. Consideration should be given to making, say, landings and access
strips from solid plate at intervals to assist in promoting a feeling of security among
users. It is also in common use externally due to its excellent performance in wet
weather.
Concrete floors are in use where heavy-duty non-removable floor areas are necessary. It is not normally advisable to use beam-and-pot-type floorings in any heavy
industrial environment due to the damage that can be suffered by lightweight thinwalled blocks. For similar reasons, where precast concrete floors require a topping
for finishes or to act as a structural diaphragm, it is advisable to use a fully-bonded
small aggregate concrete topping with continuous mesh reinforcement, with a
typical minimum thickness of 75–100 mm.
A particular advantage of both precast flooring and metal deck permanent formwork is that in many industrial structures the floor zones are irregular in plan and
elevation and therefore cheap repetitive formwork is often not practicable.
Even where formwork can sensibly be used, the early installation of major plant
items as steel frame erection proceeds can complicate in situ concrete formwork.
Floors must be able to accept holes, openings and plant penetrations on a random
layout, and often must accept them very late in the design stage or as an alteration
after construction. This provides significant problems for certain flooring, particularly precast concrete. In situ concrete can accept in a convenient manner most types
of openings prior to construction, but it may be prudent deliberately to allow for
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Anatomy of structure
107
randomly positioned holes up to a certain size by oversizing reinforcement in both
directions to act as trimming around holes within the specified units without extra
reinforcement.
Metal deck formwork is not so adaptable as regards large openings and penetrations, since it is usually one-way spanning, especially where the formwork is of a
type that can also act as reinforcement. If there is sufficient depth of concrete above
the top of the metal deck profile, then conventional reinforcing bars can be used to
trim openings. Many designers use bar reinforcement as a matter of course with
metal deck formwork to overcome this problem and also to overcome fire protection problems that sometimes occur with unprotected metal decks used as
reinforcement.
Steel-plate flooring should be designed to span two ways where possible, adding
to its flexibility in coping with openings since it can be altered to span in one direction locally if required. Open-grid flooring is less adaptable in this respect as it only
spans one way and therefore openings will usually need special trimming and
support steelwork. Early agreement with plant and services engineers is vital to
establish the likely maximum random opening required, and any structurecontrolled restraint on location.
Other policy matters that should be agreed at an early stage are the treatment of
edges of holes, edges of floor areas, transition treatments between different floor
constructions and plant plinth or foundation requirements.
This last item is extremely important as many plant items have fixings or bearings directly on to steel and the exact interface details and limit of supply of structural steelwork must be agreed. Tolerances of erected structural steelwork are
sometimes much larger than anticipated by plant and equipment designers and
some method of local adjustment in both position and level must normally be provided. Where plant sits on to areas of concrete flooring then plinths are usually provided to raise equipment above floor level for access and pipe or cable connections.
It is convenient to cast plinths later than the main floor, but adequate connection,
for example by means of dowelled vertical bars and a scabbled or hacked surface,
should always be provided. It is usually more practical to drill and fix subsequently
all dowel bars, anchor bolts for small-scale steelwork and for holding-down plant
items, and similar fixings, than to attempt to cast them into the concrete floor.
Except in particularly aggressive environments, floor areas are usually left unfinished in industrial structures. For concrete floors hard-trowelled finishes, floated finishes and ground surfaces are all used: selection depends on the use and wear that
will occur. Steel plate (solid or open-grid) is normally supplied with either paint or
hot-dip galvanized finishes depending on the corrosiveness of the environment;
further guidance is given in BS 4592: Part 11 and by floor plate manufacturers.
Steel-plate floors are fixed to supporting steelwork by countersunk set screws, by
countersunk bolts where access to nuts on the underside is practicable for removal
of plates or by welding where plates are permanent features and unlikely to require
replacement following damage.
Alternatively, proprietary clip fixings can be used for open-grid flooring plates to
secure the plates to the underside of support beam or joist flanges.
Steel Designers' Manual - 6th Edition (2003)
108
Industrial steelwork
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
3.2.4 Main and secondary beams
The plan arrangement of main and secondary beams in industrial steelwork structures normally follows both from the layout of the main items of plant and from
the column locations. Thus a sequence of design decisions often occurs in which
main beam locations dictate the column locations and not vice versa. If major plant
occurs at more than one level then some compromise on column position and hence
beam layout may be needed.
Since large plant items normally impose a line or point loading there are clear
advantages in placing main or secondary beams directly below plant support positions. Brackets, plinths or bearings may be fitted directly to steelwork, and for major
items of plant this is preferable to allowing the plant to sit on a concrete or steel
floor. Where plant or machinery requires a local floor zone around its perimeter for
access or servicing, it is common practice to leave out the flooring below the plant
for access or because the plant protrudes below the support level.
Deflection requirements between support points should be ascertained. They may
well control the beam design since stringent limits, for example relative deflections
of 1 in 1000 of support spans, may apply. In addition when piped services are connected to the plant then total deflections of the support structure relative to the
beam-to-column intersections may also need to be limited. Relative deflections can
best be controlled by the use of deep beams in (lower-grade steel if necessary), and
total deflections by placing columns as close as possible to the support positions.
It is preferable to avoid the necessity for load-hearing stiffeners at support points
unless the plant dimensions are fixed before steelwork design and detailing take
place. Where this is not possible then stiffened zones to prevent secondary bending
of top (or bottom if supports are hung) flanges should be provided, even if the design
requirements do not require load-bearing stiffeners. This then allows a measure of
tolerance for aligning the support positions without causing local overstressing
problems.
The stiffness of major plant items should be considered, at least qualitatively, in
the steelwork design. Deep-walled tanks, bunkers or silos for example may well be
an order of magnitude stiffer than the steel supporting structure.
The loading distribution given by the plant design engineers will automatically
assume fully stiff (zero deflection) supports. When the stiffness of the supporting
structure is not uniform in relation to the support point locations and loadings, then
significant redistribution of loads can take place as the structure deflects. Where the
plant support positions and loads and the structural steel layout are symmetrical
then engineering judgement can be applied without quantitative evaluation. In
extreme cases, however, a plant–structure interaction analysis may be required to
establish the loadings accurately.
When hanger supports are required then pairs of beams or channels are a convenient solution which allows for random hanger positioning in the longitudinal
direction (Fig. 3.7). Many hanger supports have springs bearings to minimize variations in support conditions due to plant temperature changes or to avoid the plant
stiffness interactions described above.
Steel Designers' Manual - 6th Edition (2003)
Anatomy of structure
109
nut
top bearing
spring
_________
RSC
____________ clearance hole in
bearing plate
__________
—
RSC
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
_____________ spacers to maintain
gap
________
hanger rod
load
Fig. 3.7
Detail at hanger support
The method of installation or removal of major items of plant frequently requires
that beams above or, less commonly, below the plant must be designed to cater for
hoisting or jacking-up the installed plant sections. Structural designers should query
the exact method and route of plant installation to ensure that the temporary hoisting, jacking, rolling or set-down loads are catered for by the steel beam framework.
Experience suggests that these data are not provided as a matter of course, and plant
designers commonly believe that the steel structure can support these loads anywhere. Where, to ease the problems, plant installation occurs during steel erection
then the method of removing plant during the building life-span may be the most
significant temporary loading for the beam framework.
When main or secondary beams are specifically designed for infrequent but heavy
lifting operations, it is good practice to fit a lifting connection to the beam to give
positive location to the lifting position and to allow it to be marked with a safe
working load. For design, smaller load safety factors are appropriate in these
circumstances.
3.2.5 Columns
Column location in industrial buildings must be decided on practical considerations.
Although regular grid layouts are desirable in normal structures, it is sometimes
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
110
Industrial steelwork
impossible to avoid an irregular layout which reflects the major plant and equipment location, the irregular floor or walkway layouts and an envelope with irregular wall and roof profiles to suit. Obviously some degree of regularity is of
considerable benefit in standardizing as many secondary members as possible; a
common method of achieving this is to lay out the columns on a line-grid basis with
a uniform spacing between lines. This compromise will allow standard lengths for
beam or similar components in one direction while giving the facility to vary spans
and provide direct plant support at least in the other direction. The line-grids should
be set out perpendicular to the longer direction of the structure if possible.
When vertical loadings are high and the capacity of rolled sections is exceeded,
several types of built-up columns are available. Where bending capacity is also of
importance, large plate I-sections are appropriate, for example in frameworks where
rigid frame action is required in one direction. However, if high vertical loads dominate then fabricated box columns are often employed (Fig. 3.8). Design of box
columns is principally constrained by practical fabrication and erection considerations. Internal access during fabrication is usually necessary for fitting of internal
stiffeners, and similarly internal access may be needed during erection for making
splice connections between column lengths, or for beam-to-column connections.
Preferred minimum dimensions are of the order of 1 m with absolute minimum
dimensions of about 900 mm. Whenever possible, column plates should be sized
to avoid the necessity of longitudinal and transverse stiffeners to control plate
buckling. The simpler fabrication that results from the use of thick plates without
stiffeners should lead to overall economies, and the increased weight of the member
is not a serious penalty to pay in columns; the same argument does not apply to
long-span box girders where increases in self-weight may well be of overriding
importance. Under most conditions of internal exposure no paint protection to the
box interior is necessary. If erection access is needed then simple fixed ladders
should be provided. Transverse diaphragms are necessary at intervals (say 3–4 times
the minimum column dimension) to assist in maintaining a straight, untwisted
profile, and also at splices and at major beam-to-column intersections even if, as is
usually the case, rigid connections are not being used. Diaphragms should be welded
to all four box sides and be provided with manhole cut-outs if internal access is
needed.
3.2.6 Connections
Connections between structural elements are similar to those in general structural
practice. Specific requirements relating to industrial steel structures are considered
here.
On occasions industrial plant and equipment may impose significant load variations on the structure and so consideration must be given to possible fatigue effects.
Since basic steel members themselves are not susceptible to fatigue failure in normal
Steel Designers' Manual - 6th Edition (2003)
Anatomy of structure
internal
stiffener
internal
base
stiffener
111
splice
=
=
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
mm. 900
preferred 1000
single V
butt
bolt
.—machined
single v/
butt\
I
section D—D
D
nternal
bolts
plate
section A—A
Fig. 3.8
section B—B
machined
section c—c
Typical details of box columns
conditions, attention must be focused on fatigue-susceptible details, particularly
those relating to welded and other connections. Specific guidance for certain types
of structure is available, and where this is not relevant, general fatigue design
guidance can be used.
Of general significance is the question of vibration and the possible damage to
bolted connections that this can cause. It should be common practice for steelwork
in close contact with any moving machinery to have vibration-resistant fixings. For
main steelwork connections there is a choice between using HSFG bolts, which are
inherently vibration-resistant, or using normal bolts with lock-nuts or lock-washer
systems. A wide variety of locking systems is available which can be selected after
consultation with the various manufacturers.
Steel Designers' Manual - 6th Edition (2003)
Industrial steelwork
112
plate girder
lood—bearing
stiffener
\actuaIwidth
bearing
detailed section
of rocker
bearing
area
(shaded)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
top
bottom
location
keep
plates
plan on cap plate
Fig. 3.9
Rocker bearing – plate girder to box column
Connection design for normally-sized members should not vary from established
practice, but for the large box and plate I-section members that are used in major
industrial steel structures, connections must be designed to suit both the member
type and the design assumptions about the joints. For particularly deep beam
members, where plate girders are several times deeper than the column dimensions,
assumed pin or simple connections must be carefully detailed to prevent inadvertent moment capacity. If this care is not taken, significant moments can be introduced into column members even by notional simple connections due to the relative
scale of the beam depth.
In certain cases it will be necessary to load a column centrally to restrict bending
on it: a typical example is where deep suspension girders on power station boilers
apply very high vertical loadings to their supporting columns. Here, a rocker cap
plate detail is often used to assure centroidal load transfer into the column (Fig.
3.9). Conventional connections on smaller-scale members would not usually require
such a precise connection as load eccentricities would be allowed for in the design.
3.2.7 Bracing, stiff walls or cores
Section 3.2.2 discusses the particular features of industrial steel structures in relation to achieving a horizontal or sway load path and describes the various methods
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Anatomy of structure
113
by which horizontal loads can be satisfactorily transferred to braced bays or other
vertical stiff elements. General design requirements and some practical suggestions
are also given in section 3.2.2 for braced steel bay design.
The layout on plan of vertically stiff elements is frequently difficult even in conventional and regularly framed structures. General guiding principles are that the
centre of resistance of the bracing system in any direction should be coincident with
the centre of action of the horizontal forces in that direction. In practice this means
that the actions and resistances should be evaluated, initially qualitatively, in the
two directions perpendicular to the structural frame layout.
Another desirable feature which is also common to many structures is that the
braced bays or stiff cores should be located centrally on the plan rather than at the
extremities. This is to allow for expansion and contraction of the structure without
undue restraint from the stiff bays, and applies equally to a single structure or to an
independent part of a structure separated by movement joints from other parts. It
is difficult to achieve this ideal in a regular and uniform structure, and almost impossible in a typically highly irregular industrial steelwork structure. Fortunately, steelwork buildings, particularly those without extensive reinforced concrete floors and
with lightweight cladding, are extremely tolerant of temperature movements and
rarely suffer distress from what may be considered to be a less than ideal stiff bay
layout.
The procedure for design purposes should be as follows. First a basic means of
transferring horizontal loading to foundation level must be decided, and guidance
on this is given in section 3.2.2. In either or both directions, where discrete braced
bays or stiff walls or cores are being utilized, then initially a geometric apportionment of the total loading should be made by an imaginary division of the structure
on plan into sections that terminate centrally between the vertically stiff structural
element. The loadings thus obtained are used to design each stiff element.
When this process has been completed and if the means exist, by horizontal
diaphragm or adequate plan bracing, to force equal horizontal deflections on to each
element, a second-stage appraisal may be needed to investigate the relative stiffness of each stiff element. Then the horizontal loading can be distributed between
vertical stiff elements on a more accurate basis and the step process repeated.
Considerable judgement can be applied to this procedure, since it is usually only
of significance when fundamentally differing stiff elements are used together in one
direction on the same structure. For example, where a horizontal diaphragm or plan
bracing exists and where some of the stiff elements are braced steelwork and some
are rigid frames, it will normally be found that the braced frames are relatively stiffer
and will therefore carry proportionately more load than the rigid frames. Similarly,
where a combination of concrete shear walls (or cores) and braced frames is used,
then a relative stiffness distribution will be needed if an effective horizontal
diaphragm or plan bracing ensures sensibly constant horizontal deflection or sway
(Fig. 3.10).
Steel Designers' Manual - 6th Edition (2003)
114
Industrial steelwork
w uniformly distributed horizontal load
id I
initial: distribute loading by geometry
0.35w
0.35w
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
0.0875w O.0875w
0.125w
accurate: stiff bracing attracts majority of loading
vertical bracing (VB)
MV moment frame
(assume VB has 4 x stiffness of MV)
Fig. 3.10
Apportioning horizontal loads to vertical stiff elements
3.3 Loading
3.3.1 General
Two difficulties exist in defining the appropriate loadings on industrial buildings.
First, the actual weight, and particularly the details of the position and method of
load application, of items of plant or equipment must be established. Even for
routine or replicated plant this information is difficult to obtain in a form that suits
the structural engineer. When the plant or equipment is being custom-built then the
problem becomes one of timing: information from the plant designer may not be
available early enough for the structural design.
The second difficulty that frequently occurs is in the choice of a general, uniformly
distributed imposed loading for any remaining floor areas not occupied by items of
plant or equipment. Guidance from codes of practice must be used carefully when
it is applied only to circulation spaces between all the fixed items of plant which are
Steel Designers' Manual - 6th Edition (2003)
Loading
115
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
known and whose loading has been evaluated and is considered separately as dead
load. Further advice on both these problems is given in section 3.3.2.
The structural steelwork designer must take an open-minded approach to the
loading information. It must be appreciated that early information from the plant
designers will represent only estimated loadings, and an uncertainty allowance may
well already be included in the values supplied, so that further allowances may be
unnecessary. However, it frequently turns out that many secondary plant items are
added at a later stage in the information process. The loadings from these should
be satisfactorily absorbed into an initial uniformly distributed loading allowance
provided that a logical scheme for dealing with loading is established at an early
stage.
3.3.2 Process plant and equipment
The most important advice to the structural designer of steelwork for an industrial
purpose is to ensure reasonable familiarity with the entire process or operation
involved. Existing facilities can be visited, and the plant designers and operators will
normally be more than willing to give a briefing which provides the opportunity to
describe the form of structure envisaged to the plant designer at an early stage so
avoiding later misunderstandings.
Some attempt should also be made to understand the jargon of the industry in
question in order to gain the confidence of the plant designers and to allow effective intercommunication.
Particular examples of structural requirements which are not easily recognized as
important by plant designers, and which should therefore be fully explained, are as
follows:
(1) The physical space requirements of bracing members and the fact that they
cannot be moved locally to give clearances.
(2) The actual size of finished steelwork taking account of splice plates, bolt heads
and fittings projecting from the section sizes noted on drawings.
(3) The fact that a steel structure is not 100% stiff and that all loads cause
deflections.
(4) The fact that a steel structure may interact with a dynamic loading and that
dynamic overload multipliers calculated or allowed on the assumption of a fully
rigid or infinite mass support are not always appropriate.
The basis of loading information for plant and equipment must be critically examined. Frequently loadings are given as a single all-up value, the components of which
may not all act together or have only a very small probability of so doing. Alternatively, the maximum loadings may represent a peak testing condition or a fault, overload condition, whereas normal operating loadings may be considerably less in
value. Worthwhile and justifiable savings in steelwork can be made if a statistically
Steel Designers' Manual - 6th Edition (2003)
116
Industrial steelwork
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
based examination of the expected frequency and duration of such unusual conditions is made, leading to the adoption of reduced load factors without reducing the
overall factor of safety. Comparisons with, for example, wind loading can be used
to establish on a reasonably logical basis the appropriate load factors.
By definition, items of fixed plant can be treated as dead load in accordance
with BS 6399: Part 12 when specific location and loads are known. At the early
stages of design it is usual to adopt a relatively large imposed loading which will
have to cater for fixed items of plant or equipment, the existence of which may not
even be known at this stage and certainly not the location and loading data. The
choice of what imposed loading to use at this stage can be assisted by the following
guidance:
very light industrial processes
medium/average industrial processes
very heavy industrial processes
7.5 kN/m2
10–15 kN/m2
20–30 kN/m2
One factor which will influence the choice within these values is the timing of the
release of final plant and equipment design data in relation to the steelwork design
and fabrication detailing process. If a second-stage design is possible then a lower
imposed load can be allowed since local variations needed to account for specific
items of fixed plant which exceed the imposed plan loading allowance can be
accommodated.
Under these conditions it can also be worthwhile, particularly for designers with
previous experience of the industrial process, to design columns and foundations
for a lower imposed loading than beams. In certain layouts with long span main
beams or girders at wide spacings this preliminary reduction can also be used
for these members. It should be stressed that these proposals are not intended to
contradict or override the particular reduced loading clauses in BS 6399,2 but are a
practical suggestion for the preliminary design stage.
When detailed plant layouts with location and loading data become available,
fixed plant and equipment can be considered as dead load and subject therefore to
the appropriate load factor, 1.4 instead of 1.6. Remaining zones of floor space
without major items of plant should be allocated an imposed load which should
reflect only the access and potential use of the floor and may therefore very well be
reduced from the preliminary imposed loading, often in the range of 5–10 kN/m2.
Specific laydown areas for removal, replacement or maintenance of heavy plant
are the only likely exceptions to this range of loading.
An alternative scheme for dealing with the second stage of loading information
is to institute a checking procedure where the equivalent loading intensity of plant
and equipment is calculated for each item as follows:
1.4
weight (kN)
¥
1.6 plan area (m 2 )
the factor 1.4/1.6 being introduced to cater for the reclassification from imposed
load to dead load.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Loading
117
Only in bays where this equivalent loading intensity exceeds the preliminary
imposed loading will further evaluation be required, unless a scheme of reduced
imposed loading for long-span beams and columns has been adopted, in which case
a rigorous check on actual loading intensity must be carried out.
Many items of plant contain moving parts which are always liable to exert
dynamic or vibratory loadings on to the structure. Assessing the structural response
to these loadings is never easy, particularly when, as is usually the case, the plant
and the structure are being designed out of sequence and by different designers.
Many rule-of-thumb approximations exist, in which dynamic effects are allowed
for by percentage increases to dead load. Some instances of this are codified: for
example, crane vertical loadings are factored by 1.25 (see BS 6399: Part 12). Similar
multipliers to static loading can be provided by plant suppliers but these must be
used with great care and the limitations of such an approach should be appreciated.
The dynamic load from one-off impact-type actions or from successive load applications at irregular time intervals depends both on the characteristics of the plant
itself and on the mass and stiffness of the structure to which they are applied. Any
dynamic load calculated independently from a knowledge of the structure will be
based on assumed structure properties and often on limit values such as zero mass
or infinite stiffness. Provided that the structure remains elastic and that resonance
or similar frequency-related amplification does not occur the approximate dynamic
factors normally represent upper limits. If more accurate evaluations are warranted
for major items of plant, a plant–structure interaction dynamic analysis must be
undertaken.
Vibrating plant or equipment will transmit vibrations to the structure, the effects
of which can be dramatic. The frequency of vibration of the plant can induce resonance in the structure if the natural frequency of vibration of the whole or any part
of the structure subjected to the vibrating force is equal or very close to the forcing
frequency. Considerable judgement is needed to identify all possible modes of
natural vibration that could be excited by the vibrating equipment or machinery.
To safeguard against a resonant response there should be no natural frequency of
vibration within the range of 0.5–1.5 times the forcing frequency. Useful guidance
on calculating natural frequencies of structures is given by Bolton.3
A useful initial step is to calculate the lowest natural frequency of vertical motion
of the floor construction and to ensure that it is higher than 3–4 Hz (cycles per
second). This should avoid problems of response to human-induced vibration, the
so-called ‘springiness’ of floors, caused by resonant amplification of footfalls, which
lie in the range of 1–2 Hz. Then only plant-induced vibrations of lower than 1 Hz or
higher than 2 Hz need to be investigated. Anti-vibration mountings for plant can be
specified but these are not an automatic success as they only filter the induced forces
and some form of variable loading will still be transmitted to the structure. Various
types and grades of anti-vibration mountings are available; information on their
suitability can be obtained from the manufacturers, based on details of both the item
of plant and the structure.
Many of these devices are intended to reduce vibration of the plant or equipment
itself. Care is required to ensure that the altered forcing vibrations which are
Steel Designers' Manual - 6th Edition (2003)
118
Industrial steelwork
imposed on to the structure do not have a secondary adverse effect of inducing
resonant vibrations into it.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
3.3.3 Lateral loadings from plant
Lateral loadings imposed by plant on the structure derive from three sources. These
are considered separately although there is a common theme throughout that the
operating process undergoes a change in regime which is the cause of loading. Many
of the actions which give rise to horizontal loads also cause vertical loads or at least
vertical loading components which must be incorporated into the design. However,
whereas vertical loadings are usually readily understood and allowed for in loadings provided by plant designers, one feature of the horizontal components of such
loadings that causes confusion is the fundamental concept of equilibrium. Notwithstanding exotic situations where masses (projectiles) leave or impinge on a structure, or where motion energy is dissipated as heat of friction, equilibrium
considerations dictate that lateral components of forces are in balance and consequentially they are often ignored. This is not satisfactory as balancing components
may act a considerable distance apart (the load path must be examined in detail)
or indeed the balancing components may act at different levels, leading to a more
conventionally understood requirement to transfer lateral loading.
The three causes are as follows:
(1) temperature-induced restraints
(2) restraint against rotational or (more rarely) linear motion
(3) restraint against hydraulic or gaseous pressures.
Where plant undergoes a significant change in temperature, plant designers will
typically assume that the structure is fully rigid and so can absorb the forces generated by application of restraints at the structure connection points. They will then
design the plant itself for the additional stresses that are caused by preventing free
thermal expansion or contraction. This is a safe upper bound procedure since the
forces generated in both structure and plant represent maxima, with any deflection
at the support reducing forces in both elements. Naturally the structural steelwork
designer must be made aware both of the assumption of zero deformation so that
the support can be made as stiff as possible in the required direction, and of the
forces that are thus imposed.
In spite of the apparent complications of this approach, it is frequently adopted
by plant designers for convenience on small items, and to avoid complexities in
interconnection between plant items and with piped and ducted services on larger
items.
The alternative approach, common on major plant subjected to significant
thermal variation such as boilers and ovens, is to assume completely free supports
with zero restraint against expansion or contraction. This is a lower bound solution
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Loading
119
which needs some rational assessment of possible forces that could result from
bearing or guide misalignment, malfunction or simple inefficiency.Where large plant
items are involved the forces even at such guides or bearings can be significant if
the balancing reaction is, for example, at ground level or even outside the structure
itself.
The lateral forces from constant speed, rotating machinery, which are usually
relatively low, are generally balanced by an essentially equal and opposite set of
forces coming from the source of motive power. Nevertheless, their point of application must be considered and a load path established which transfers either them
back to balance each other or alternatively down to foundation level in the conventional way. The structural steel designer must have a completely clear and unambiguous understanding of the source and effect of all the moving plant forces, to
ensure that all of them are accounted for in the crucial interface between plant and
equipment.
The start-up forces when inertia of the plant mass is being overcome, and the fault
or ‘jamming’ loads which can apply when rotating or linear machinery is brought
to a rapid halt, need consideration. They are obviously of short duration and so can
justifiably be treated as special cases with a lower factor of safety. At the same time
the load combinations that can actually co-exist should be established to avoid any
loss of economy in design.
Pressure pipe loadings, significant in many plant installations, occur wherever
there are changes in direction of pipework and associated pipework supports or
restraints. Thermal change must also be considered. Pipework designers may well
combine the effects to provide a schedule of the total forces acting at each support
position. Since pipework forces are normally not reversible in direction, then on
major installations the pipework designer may wish to make the installation by
‘forcing’ the pipework configuration, deliberately making sections too short or too
long and then prestressing the pipework so that when installation is complete the
operating conditions take the internal pipework stresses through a neutral stress
zone and then reverse them.
Forcing is a complex procedure sensitive to lack of fit at the pipework supports
and restraints and often to other factors such as the ambient temperature during
installation and the exact sequence of connections. Where major pipe installations
are planned and where the pipework designers are adopting these techniques, the
structural steelwork designer should acknowledge that the loading values quoted
may not be achieved in practice and make a further allowance.
3.3.4 Wind loadings
Wind loadings on fully enclosed industrial structures do not differ from wind loadings on conventional structures. The only special consideration that must be given
applies to the assessment of pressure or force coefficients on irregular or unusualshaped buildings. A number of sources give guidance on this topic, and specific
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
120
Industrial steelwork
advice can be sought from the Building Research Establishment (BRE) Advisory
Service or other specialist advisors.
However, on partly or wholly open structures with exposed plant or equipment
great care must be exercised in dealing with wind loading.
It is frequently the case that the total wind loads are higher than on a fully clad
building of the same size, due to two causes. First, small structural elements attract
a higher force than equivalent exposed areas which form part of a large façade.
Secondly, repetitive structural elements of plant items which are nominally shielded
by any particular wind direction are not actually shielded and each element is subjected individually to a wind load. The procedure for carrying out this assessment
is not covered in detail in BS 6399: Part 24 but guidance is available from the references listed in that document.
Loading on particularly large individual pieces of plant or equipment exposed to
wind can be calculated by considering them to be small buildings and deriving
overall force coefficients that relate to their size.
For smaller or more complex shapes, such as ductwork, conveyors and individual
smaller plant items, it is sensible to take a conservative and easy to apply rule-ofthumb and use a net pressure coefficient Cp = 2.0 applied to the projected exposed
area. The point of application of wind loadings from plant items on to the structure
may be different from the vertical loading transfer points if sliding bearings or
guides are being used.
3.3.5 Blast loadings
This section deals only with blast loadings from industrial processes and not with
any generalized design requirements to survive blast loadings from unspecified
sources or of unspecified values.
Varying requirements exist for blast loadings. Typical examples are:
(1) transformers, where the requirement is usually to deflect any blast away from
other vulnerable pieces of plant but where frequently one or more walls and
the roof are open, serving to dissipate much of the energy discharged
(2) dust or fine particle enclosures, which are often wholly inside enclosed
buildings.
The decisions to be made are as follows.
(1) Can the potential source of the blast be relocated outside the building
altogether, in a separate enclosure?
(2) If not, can it be placed against the external wall with arrangements to have a
major permanent vented area or a specially designed blow-off panel, both of
which will limit loadings on the remaining structure?
(3) Where the location cannot be controlled, it must be established which direction
Steel Designers' Manual - 6th Edition (2003)
Loading
121
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
or directions require full protection against damage and which can tolerate
certain degrees of damage.
Loading data given by plant designers are usually stated in terms of peak pressures to be applied to projected areas in line with the potential source of the blast.
The validity of the data must be treated as being highly suspect since blast loadings
are classic examples of true dynamic loading where the time-dependent response
of the structure actually determines the loading that is imposed. Quoted blast pressures, which are probably derived from theoretical considerations of high-rigidity
high-mass targets fully enclosing the source, may be invalid for steel structures
which have tremendous capacity to deform rapidly, absorbing energy and thereby
reducing and smoothing out peak blast pressures. Whatever results are obtained
from analysis or calculation, it is good practice to use a grade of steel which has
good ductility, a lower yield stress from which to commence ductile behaviour and
a long and reliable extensibility prior to fracture. While this will ensure reasonable
material behaviour, overall ductility of the structure also depends on stability
against premature buckling and the ductile behaviour of connections.
Steelwork designers should acknowledge the very imprecise nature of most blast
loading data, even when the potential source of the blast is precisely located and
specified. They should thus direct their attention to ensuring that collapse does
not occur until major deflections and rotations have occurred, following normal
guidelines for achieving plastic behaviour.
3.3.6 Thermal effects
This section deals with thermal effects from environmental factors; specific consideration of plant-induced thermal effects is given in section 3.3.3. Conventional guidance on the provision of structural expansion or contraction joints is often
inappropriate and impractical to implement, and indeed joints frequently fail to
perform as intended.
The key to avoiding damage or problems from thermal movements is to consider
carefully the detailing of vulnerable finishes (for example, brickwork, blockwork,
concrete floors, large glazing areas and similar rigid or brittle materials). Provided
that conventional guidance is followed in the movement provisions for these
materials, then structural joints in steel frames can usually be avoided unless there
is particularly severe restraint between foundations and low-level steelwork.
Many industrial structures with horizontal dimensions of 100–200 m or more have
been constructed without thermal movement joints, usually with lightweight, nonbrittle cladding and roofing, and a lack of continuous suspended concrete floors.
When high restraint close to foundations or vulnerable plant or finishes are
present, a thermal analysis can be carried out on the steel framework to examine
the induced stresses and deflections and to evaluate options such as the introduction of joints or altering the structural restraints. Where steelwork is externally
Steel Designers' Manual - 6th Edition (2003)
122
Industrial steelwork
exposed in the UK, the conditions vary locally but a minimum of -5°C to +35°C
suffices for an initial sensitivity study. For many of the likely erection conditions a
median temperature of 15°C can be assumed, and a range of ±20°C can thus be
examined. Where the effects of initial investigations based on these values highlight
a potential problem, more specific consideration can be given to the actual characteristic minimum and maximum temperatures (advice in the UK is available from
the Meteorological Office), and steps may need to be taken to control erection,
particularly foundation fixing, to take place at specified median temperatures.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
3.4 Structure in its wider context
Industrial structural steelwork is inherently inflexible, being purpose-designed for a
particular function or process, and indeed often being detailed to suit quite specific
items of major plant and equipment. Nevertheless, it is important to try to cater for
at least local flexibility to allow minor alterations in layout, upgrading or replacement of plant items. The most appropriate way to ensure this is to repeat the advice
that has been given on numerous occasions already in this chapter. The designer
must understand the industrial process involved and be aware of both structural and
layout solutions that have been adopted elsewhere for similar processes. Previous
structural solutions may not be right, but it is preferable to be aware of them and
positively to reject them for a logical reason, than to reinvent the wheel at regular
intervals.
General robustness in industrial buildings may be difficult to achieve by the
normal route of adopting simple, logical shapes and structural forms, with welldefined load-paths and frequent effective bracing or other stability provisions.
Instead of these provisions, then, it is sensible to ensure that a reasonable margin
exists on element and connection design. Typically, planning for a 60–80% capacity
utilization at the initial design stages will be appropriate, so that even when these
allowances are reduced, as so frequently occurs, during the final design and checking stages, adequate spare capacity still exists to ensure that no individual element
or joint can weaken disproportionately the overall structural strength of the
building.
References to Chapter 3
1. British Standards Institution (1995) Industrial type metal flooring, walkways and
stair treads. Part 1: Specification for open bar gratings. BS 4592, BSI, London.
2. British Standards Institution (1996) Loading for buildings. Part 1: Code of
practice for dead and imposed loads. BS 6399, BSI, London.
3. Bolton A. (1978) Natural frequencies of structures for designers. The Structural
Engineer, 56A, No. 9 Sept., 245–53.
Steel Designers' Manual - 6th Edition (2003)
Further reading
123
4. British Standards Institution (1995) Loading for buildings. Part 2: Code of practice for wind loads BS 6399, BSI, London.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Further reading for Chapter 3
Booth E.D., Pappin J.W. & Evans J.J.B. (1988) Computer aided analysis methods for
the design of earthquake resistant structures – a review. Proc. Instn Civ. Engrs,
84, Part 1, Aug., 671–91.
Fisher J.M. & Buckner D.R. (1979) Light and Heavy Industrial Buildings. American
Institute of Steel Construction, Chicago, USA, Sept.
Forzey E.J. & Prescott N.J. (1989) Crane supporting girders in BS 15 – a general
review. The Structural Engineer, 67, No. 11, 6th June, 205–15.
Jordan G.W. & Mann A.P. (1990) THORP receipt and storage – design and construction. The Structural Engineer, 68, No. 1, 9th Jan., 7–13.
Kuwamura H. & Hanzawa M. (1987) Inspection and repair of fatigue cracks in crane
runway girders. J. Struct. Engng, ASCE, 113, No. 11, Nov., 2181–95.
Mann A.P. & Brotton D.M. (1989) The design and construction of large steel framed
buildings. Proc. Second East Asia Pacific Conference on Structural Engineering
and Construction, Chaing Mai, Thailand, 2, Jan., 1342–7.
Morris L.J. (Ed.) (1983) Instability and plastic collapse of steel structures. Proc. M.R.
Horne Conference. Granada, St Albans.
Taggart R. (1986) Structural steelwork fabrication. The Structural Engineer, 64A,
No. 8, Aug., 207–11.
Steel Designers' Manual - 6th Edition (2003)
Chapter 4
Bridges
by ALAN HAYWARD
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
4.1 Introduction
Use of structural steel in bridges exploits its advantageous properties of economically carrying heavy loads over long spans with the minimum dead weight. Steel is
however suitable for all span ranges, categorized in Table 4.1.
For long spans steel has been the natural solution since 1890 when the Firth of
Forth cantilever railway bridge, the world’s first major steel bridge, was completed.
For short and medium spans concrete bridges held a monopoly from 1950 to 1980
because of the introduction of prestressing and precasting. Developments in steel
during this period such as higher tensile strength and improved welding techniques
were applied mainly to long spans. However, improvements in construction methods
from 1980 have enabled steel to improve its market share within Europe and other
continents to more than 50% for short and medium spans. Contributing factors to
this trend are shown in Table 4.2.
Where traffic disruption during construction must be minimized then steel is
always suitable. Most bridgework is now carried out under these conditions so
that structural steel will continue as a primary choice. Steel has an advantage where
speed of construction is vital; it is no coincidence that this usually results in cost
economies. If rapidly erected steelwork is used as a skeleton from which the slab
and finishes can be carried out without need for falsework then the advantages
(summarized in Table 4.3) are fully realized.
For short and medium span highway bridges composite deck construction is economic because the slab contributes to the capacity of the primary members.1 For
continuous spans it uses the attributes of steel and concrete to best advantage. In
cases where construction depth is restricted, for example in developed areas, then
half-through or through construction is convenient; this is common for railway and
pedestrian bridges.
For long spans, including suspension or cable-stayed bridges, all-steel orthotropic
plate floors are used. Although the intrinsic costs of a steel orthotropic plate are
higher (often up to about four times more) than an equivalent concrete slab, the
advantage in dead weight reduction (approximately 1 : 3 ratio) may more than offset
this when the overall economy is considered. Steel decks are also employed when
erection must be completed in limited occupations, such as for railway bridges on
existing routes. Movable bridges of swing, lift, rolling-lift (‘bascule’) or retractable
type usually employ steel floors to rationalize the amount of counterweighting and
124
Steel Designers' Manual - 6th Edition (2003)
Introduction
125
Table 4.1 Span ranges of bridges
Short
Medium
Long
Up to 30 m
30 m to 80 m
80 m
Table 4.2 Factors contributing to the improved market share for steel in short and medium spans
from 1980
Factor
Reasons
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(1) Stability in price of rolled steel products
(2) Automation of fabrication processes
(3) Faster erection with larger components
(4) Improved design codes
(5) Evidence of durability problems with concrete
(6) Better education in steel design
Stability worldwide
Mechanized equipment for material preparation,
welding and girder fabrication
Availability of high-capacity cranes. Delivery in
longer lengths
Easier yet more rigorous methods of design
Life expiry of concrete bridges under 30 years
old
SCI and BCSA publications and training courses
Table 4.3 Advantages of steel bridges
Feature
(1) Low weight of deck
. . . leading to . . .
Advantages
Economy
(2) Light units for erection
Smaller foundations. Typical
30–50% reduction of weight
compared with concrete decks.
Erection by mobile cranes
(3) Bolted site connections
Minimal site inspection
Flexible site planning
(4) Prefabrication in factory
Effective quality control
More reliable product
(5) Modern methods of
protective treatment
Application of treatment before
erection. Long life to first
maintenance.
Predictable whole life cost
of structure
(6) Use of weathering steel
Rapid construction
Minimum whole life cost
(7) Shallow construction depth
Minimum length of approach
grades
Overall economy of highway.
Slender appearance.
(8) Self-supporting steel
Elimination of falsework
Easier construction, especially
for high structures
(9) Continuous spans
Fewer bearings and joints
Slender appearance.
Reduced maintenance cost.
the mechanical equipment. Costs of this equipment usually exceed that of the bridge
superstructure. For long spans of suspension or cabled-stayed form then special
considerations affect the design including aerodynamic behaviour, the feasibility
of deep foundations in estuarial conditions, cables with anchorages, non-linear struc-
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
126
Bridges
tural behaviour and the absolute necessity to include the effects of the erection procedure in the design process.
This chapter on bridges gives emphasis to initial design, an important stage of the
process, because the basic decisions as to member proportions, spacings and splice
positions vitally affect economy of the structure. It is essential that the detailed
analysis is based upon optimized sizes which are as accurate as possible. If this is
not achieved then the detailed design will be inefficient because time consuming
repetitive work will have been expended, adversely affecting the economy of the
design and the construction costs. Guidance is given on the initial design of highway
bridges using composite deck construction, which is a significant proportion of the
number of steel bridges built, although since 1990 a significant number of railway
bridges have been constructed for new-build schemes such as the Channel Tunnel
Rail Link and in replacement of life expired structures.
Steel is particularly suitable for the strengthening and repair of existing bridges
arising from increases in highway traffic loadings and the incidence of accidental
impact damage from road vehicles. Steel is suitable for such work using welded or
bolted strengthening.
4.2 Selection of span
The majority of bridges fall within the category of short span because for many
crossings of rivers, railways or secondary highways a single span of less than 30 m is
sufficient. For multiple-span viaducts a decision on span length must be made, which
depends on factors shown in Table 4.4.
For long viaducts it is necessary to carry out comparative estimates for different
spans to determine the optimum choice, as shown in Fig. 4.1.
Table 4.4 Factors which decide choice of span for viaducts
Factor
Reasons
Location of obstacles
Pier positions are often dictated by rivers, railway tracks and
buried services
Construction depth
Span length may be limited by the maximum available
construction depth
Relative superstructure and
substructure costs
Poor ground conditions require expensive foundations;
economy favours longer spans
Feasibility of constructing
intermediate piers in river
crossings
(a) Tidal or fast-flowing rivers may preclude intermediate piers
(b) For navigable waterways, accidental ship impact may
preclude mid-river piers
Height of deck above ground
Where the height exceeds about 15 m, costs of piers are
significant, encouraging longer spans
Loading
Heavier loadings such as railways encourage
shorter spans
Steel Designers' Manual - 6th Edition (2003)
Selection of type
127
1000
TTTI
500
optimum span
10
20
30
40
50
60
70
span (m)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 4.1
Choice of optimum span for viaducts
Table 4.5 Typical optimum span ranges (m) for viaducts
Conditions
Simple foundations (spread footing or short piles)
Highway
Railway
25–45
20–30
Difficult foundations (piles >20 m long)
35–55
25–40
Piers >15 m high
45–65
30–45
Typical optimum spans are shown in Table 4.5.
Long spans are usually adopted only when a larger number of shorter spans are
precluded by restrictions of the site. This is because the material content and cost
per unit length of long spans is much greater. Therefore the bridging of a wide river
or estuary should generally use a number of medium spans. Only if there is a high
risk of shipping collision, or if the depth and speed of water flow is such as to make
foundation construction very difficult, should a long-span bridge be chosen.
4.3 Selection of type
Suspension or cable-stayed bridges are suitable for the longest spans, but are less
suitable to support heavy loading across short or medium spans. At the same time
medium-span footbridges can appropriately be suspension, cable-stayed or arch
types because concentrated loading is absent. Some of the considerations given to
long spans such as aerodynamics need to be applied to footbridges. For footbridges
the phenomenon of pedestrian-excited vibrations needs to be considered and tends
to affect the design of spans exceeding about 25 m. The possibility of horizontal
oscillation needs consideration if the structure is laterally flexible or is mounted
on slender supports. Other bridge types such as arches or portals may be suitable
Steel Designers' Manual - 6th Edition (2003)
128
Bridges
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
in special locations. For example, an arch is a logical solution for a medium span
across a steep-sided ravine.
Through trusses are suitable for medium spans where the available construction
depth is limited. The vast majority of short- and medium-span highway bridges are
formed with composite construction because the highway profile can be arranged
to suit the depth available.
For short and medium spans the most important factor which influences the type
is the available construction depth. This particularly affects railway bridges because
it is rarely feasible to modify existing track levels. Where depth is limited then types
such as arch, truss or half-through plate girder offer an alternative solution. Types
of steel bridge are shown in Fig. 4.2 with their normal economic span range and the
world’s longest. Each is briefly described below.
4.3.1 Suspension bridges
Suspension bridges (Fig. 4.3) are used for the longest spans across river estuaries
where intermediate piers are not feasible. The cables form catenaries supporting
both sides of the deck and are tied to the ground usually by gravity foundations
sometimes combined with rock anchors. Thus ground conditions with firm strata at
or close to the surface of the ground are essential. Towers are usually twin steel or
concrete box members which are braced together above the roadway level. They
are designed so as to be freestanding under wind loading during construction until
the cables are installed. Cables are either a compacted bundle of parallel high tensile
steel strands (commonly 5 mm diameter) installed progressively by ‘spinning’ or
may be formed from a group of wire ropes. Deck hangers are wire ropes (or round
steel rods for light loading as for a footbridge) clamped to the cable and connected
to the deck at a spacing equal to the length of each deck unit erected, typically
18 m. The construction process for suspension bridges is more time consuming than
for other types because the deck cannot be installed until the towers, anchorages,
cable and hangers are constructed.
Depending upon ground conditions, the cables can be catenaries supporting side
spans. Cables may alternatively be straight from tower top to the ground anchorages and merely support a main span, side spans being non-existent or formed as
short-span viaducts. Decks are either trusses with a steel orthotropic plate floor
spanning between or an aerofoil box girder. Footways are often cantilevered outside
the two sets of cables.
Aerodynamic behaviour must be considered in design because of the tendency
for the deck and cables to oscillate in flexure and torsion under ‘vortex shedding’
and other wind effects. This is due to the flexible nature and light weight of suspension bridges illustrated by the collapse of the USA Tacoma Narrows Bridge in
1940, which had a very flexible narrow deck consisting of twin plate girders forming
a torsionally weak deck of ‘bluff’ shape prone to wind vortex shedding. Aerodynamic considerations usually justify wind tunnel testing of models. The advantage
of an aerofoil box girder such as used on the Severn and Humber bridges is that
rn)
span
1500
Fig. 4.2
500 400
Normal span range of bridge types
1000
300 200
ioo
o
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
Selection of type
129
Steel Designers' Manual - 6th Edition (2003)
130
Bridges
main span
ground
anchorage
typical
(main span)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(a)
(b)
(c)
(d)
(e)
(f)
Fig. 4.3 Suspension bridges. (a) Three-span suspended; (b) straight back stays, viaduct side
spans; (c) straight back stays, no side spans; (d) truss deck; (e) aerofoil box deck; (f)
inclined hangers
Steel Designers' Manual - 6th Edition (2003)
Selection of type
131
it discourages vortex shedding and reduces the wind forces to be resisted by
the towers and substructures. The Severn (1966) and Bosporus Bridge (1970)
use inclined hangers that form a truss system which helps to reduce any
tendency to oscillate. Suspension bridges behave as non-linear structures under
asymmetric deck loading so that deflections may be significant. Behaviour under
such loading depends upon the combined gravity stiffness and the flexural rigidity
of the deck or stiffening girder. The type is less suitable for heavy loading such as
railway traffic, especially for short spans. Suspension bridges are sometimes suitable
for medium spans carrying pedestrian or light traffic.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
4.3.2 Cable-stayed bridges
Cable-stayed bridges (Fig. 4.4) are of a suspension form using straight cables which
are directly connected to the deck. The structure is self-anchoring and therefore less
dependent upon good foundation conditions, but the deck must be designed for
the significant axial stress from the horizontal component of the cable forces. The
construction process is quicker than for a suspension bridge because the cables and
deck are erected at the same time and the amount of temporary works is reduced.
Either twin sets of cables are used or alternatively for dual carriageways a single
plane of cables and tower can be located in the central reserve space. Two basic
forms of cable configurations are used, either ‘fan’ or ‘harp’. A fan layout minimizes
bending effects in the structure due to its better triangulation but anchorages can
be less easy to incorporate into the towers. The harp form is often preferred where
there are more than, say, four cables. The number of cables depends on the span
and cable size, which is often selected such that each fabricated length of deck (say,
20 m) contains an anchorage at one end to suit a cantilever erection method. Bridges
either have two towers and are symmetrical in elevation or have a single tower as
suited to the site.
Floors are generally an orthotropic steel plate but composite slabs can be used
for spans up to about 250 m. A box girder is essential for bridges having a single
plane of cables to achieve torsional stability, but otherwise either box girders or twin
plate girders are suitable. Aerodynamic oscillation is a much less serious problem
than with suspension bridges but must be considered. Some bridges with plate
girders incorporate non-structural aerodynamic edge fairings.
It is essential to use cables of maximum strength and modulus at a high working
stress so that sag due to self-weight, which produces non-linear effects, is negligible.
Cables are normally of parallel wires or prestretched locked coil wire rope. During
erection the cable lengths are adjusted or prestressed so as to counteract the dead
load deflections of the deck arising from extension of the cables.
4.3.3 Arch bridges
Arch bridges (Fig. 4.5) are suitable in particular site conditions. An example is
a medium single span over a ravine where an arch with spandrel columns will
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
132
Bridges
(a)
(b)
IL
IL
(d)
————
(c)
II
.1.
(e)
1
(f)
Fig. 4.4 Cable-stayed bridges. (a) Harp; (b) fan; (c) single tower; (d) single plane; (e) single
plane; (f) twin plane
Steel Designers' Manual - 6th Edition (2003)
Selection of type
133
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(a)
(b)
(c)
Fig. 4.5
Arch bridges. (a) Spandrel post arch; (b) tied arch; (c) part tied arch
efficiently carry a deck with the horizontal thrust taken directly to rock. A tied arch
is suitable for a single span where construction depth is limited and presence of
curved highway geometry or other obstruction to the approaches conflicts with the
back stays of a cable-stayed bridge.
4.3.4 Portal frame bridges
Portal frame bridges (Fig. 4.6) are mainly suitable for short or medium spans. In a
three-span form with sloping legs they can provide an economic solution by offering a reduction in span and have an attractive appearance. The risk of shipping collision with sloping legs must be considered for bridges over navigable rivers. Portal
Steel Designers' Manual - 6th Edition (2003)
134
Bridges
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(0)
(b)
Fig. 4.6
Portal frame bridges. (a) Three-span inclined legs; (b) single span
bridges tend to be less economic than beam bridges because the foundations must
be designed to resist horizontal thrust, complex details can be required at the joints
and erection is more complicated.
4.3.5 Truss bridges
Through trusses are used for medium spans where a limited construction depth precludes use of a composite deck bridge. They are suitable in flat terrain to reduce the
height and length of approach embankments and for railway bridges where existing gradients cannot be modified. A truss may be unacceptable visually; a bowstring
truss is an alternative solution. For short spans and medium spans up to 50 m, trusses
are generally less economic than plate girders because of higher fabrication cost.
They are therefore adopted only where the available construction depth is not sufficient for composite beams. (See Fig. 4.7.)
4.3.6 Girder type
Girder bridges (Fig. 4.8) predominate over the previously described types for short
and medium spans, and generally provide the most economic solution. For highway
bridges composite deck construction is generally used unless the depth is very critical in which case half-through girders or through trusses may be necessary. Railway
bridges frequently require to be of half-through or through form because of depth
limitations; these are more suitable, being generally narrower. Pedestrian bridges
Lspan
splices
Fig. 4.7 Truss bridge: (a) elevation (b) cross section (c) arrangements for sliding into
endposition
fixed
span.
link
Selection of type
120000 = 12000 x bays 10
(a)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
135
Steel Designers' Manual - 6th Edition (2003)
136
Bridges
(a)
(b)
•••••:__
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
suspended span
(c)
(d)
(e)
Fig. 4.8 Girder type bridges. (a) Single span; (b) simply-supported spans; (c) cantilever and
suspended span; (d) continuous; (e) continuous – curved soffit
are similarly suited to a half-through form so as to reduce the length of staircases
and ramps, which can often exceed the length of the span itself. Highway, railway
and pedestrian bridges of girder type for short and medium spans are further
described below.
4.3.6.1 Highway bridges – composite deck construction
(see also Chapter 17)
Composite deck construction (Fig. 4.9) should be used wherever the construction
depth will permit. If possible, multiple spans should be made continuous over the
intermediate supports, so reducing the number of bearings and expansion joints.
Continuity gives economies throughout the structure and reduces traffic disruption
arising from the maintenance needs of these vulnerable elements. A number of
options are available for maintaining continuity over intermediate supports. If
ground conditions are poor such that the predicted differential settlement of the
Steel Designers' Manual - 6th Edition (2003)
Selection of type
span
range
}
25
137
remarks
use plastic stress
analysis for
economy
may use pier
cross—head as
below
3500
--
steel cross—head
3500
3500
—
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
1RiAJ
—-—-4
00
1
II
25
to
100
shown to reduce
no. of columns
may use leaf pier
as above
multiDle PG
unsuitable curved
soil it
35
to
90
I
9000
I-
—l
suitable curved
soil it
NI <N
1<N
35
to
90
twin PG & cross girder
4600
4600
—
no bracing up to
60 m span
—
-T
35
to
100
V
1200
Fig. 4.9
7000
I
steel cross—heod
shown to reduce
no. of columns
multiple compact box
k3i
uneconomic unless
aesthetics govern
-
nrn ton trnniou1nI ho
Highway bridges – composite deck construction
no bracing up to
60 m span
35
to
100
temporary plan
bracing required
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
138
Bridges
supports is significant (say, exceeding (span/1000)) then to avoid overstress the
structure should be made statically determinate by use of simply-supported spans.
Cantilever and suspended spans are alternative options which retain some of the
advantages of continuity. A girder depth of (span/20) (girder depth excludes floor
slab) is generally economic although shallow girders can be used down to a depth
of (span/30) or less.
Rolled sections are appropriate for short spans up to 25 m span. For continuous
spans exceeding about 22 m, fabricated plate girders will show economy because
lighter flanges and webs can be inserted in the mid-span regions. Automated manufacture of plate girders means that they are highly economic when compared with
box girders. Normally a girder spacing of 2.5–3.5 m is optimum with a floor slab of
about 220–250 mm thick. Edge cantilevers should not exceed 50% of the beam
spacing and to simplify falsework should where possible be less than 1.5 m. An even
number of girders (i.e. 2,4, 6, 8, etc.) achieves better optimization for material ordering and permits girders to be braced in pairs for erection.
For medium spans exceeding 40 m where adequate construction depth is available it may be economic to use twin girders only. A number of variants are available as shown in Fig. 4.9, using a thickened haunched slab, longitudinal stringers and
cross girders to support the slab intermediately. For narrow bridges the complete
precasting of composite floors may offer advantages in speed of construction.2 Use
of girders with a curved soffit becomes economical for medium spans exceeding
45 m and efficiently achieves maximum headroom if required over the central portions of a span (see Fig. 4.9). Plate girder flanges should be proportioned so as to
be as wide as possible consistent with outstand limit to reduce the number of intermediate bracings. For practical reasons a desirable minimum flange width is about
40 mm to accommodate shear connections and to permit the possible use of permanent formwork. A maximum flange thickness of 75 mm is recommended as a
guide to avoid heavy butt welds, but thicker flanges can be used where necessary,
to 100 mm or greater.
For long spans, box girders (see Fig. 4.10) are more suitable than plate girders, for
which flange sizes would be excessive. Other reasons for using box girders for long
spans may include a need to improve aerodynamic stability, the presence of severe
plan curvature, a requirement for single column supports or very limited construction depth. However, for short and medium spans box girders are generally less economic because, although a reduction in flange sizes may be possible due to superior
load distribution properties, this is more than offset by the amount of internal
diaphragms and stiffening and the extra costs in manufacture. Fabrication costs are
higher because the assembly and welding processes are less amenable to automation than with plate girders. Also access must be permitted inside box girders for
welding, protective treatment processes, and permanent inspection. However, erection of box girders is often easier because they require minimal external bracing to
maintain overall stability. Multiple compact section box girders can be economic for
spans of up to 50 m in particular situations, and enable longitudinal stiffeners to be
eliminated. Open-top trapezoidal box girders (known as ‘bath tubs’) are widely used
in North America and possess some of the advantages of plate girders. They have
Steel Designers' Manual - 6th Edition (2003)
Selection of type
139
steel cantilevers
if slob extends
>3300
200 to 400 mm
3000 to
900 to ____________
—
3500 c/c
1200
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(a)
200 to
250 mmf
900 to
1200
_____________
—
—
(b)
220 to
250 mmf
900 to
2500 to 3500
1200
at pier
at mid-span
(c)
Fig. 4.10 Highway bridges – box girders. (a) Twin box and cross girders (spans 40–150 m);
(b) open top box (spans 40–100 m); (c) multiple box (spans 30–60 m)
Steel Designers' Manual - 6th Edition (2003)
140
Bridges
seen same use in the UK but temporary bracing is required during construction to
maintain shape and relative twist of the sections until the concrete slab is placed
and the full torsional rigidity achieved.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
4.3.6.2 Railway bridges – girder type (Fig. 4.11)
Many existing small-span railway bridges weather cast or wrought iron girders to
which rails are fixed directly without use of ballast or half-through girders with
trough floors. As these bridges have reached the end of their lives they have been
replaced in steelwork in modern form with ballasted track. The legacy of the
original decks with their very shallow construction depth has influenced modern
underline bridge practice in replacements and new bridges. The majority of railway
underline bridges therefore tend to be of half-through type. The direct fastening of
track without ballast gives the cheapest possible form of railway bridge and is appropriate for rapid transit or tramway bridges where speeds are not high.
Girder depths are generally greater than for highway bridges due to the heavier
loading and because limits for deformation are necessary. A span-to-girder depth
ratio of 12 to 15 is typical. For short spans half-through plate girders are used with
composite steel cross girders forming rigid U-frames and supporting a concrete floor
with ballasted single or double track. Through or half-through construction is appropriate for medium spans exceeding 50 m using trusses. For spans up to 39 m, the
railway authorities use a standard box girder design with a steel ribbed floor of
minimal depth which is achieved by spanning between the inner webs of trapezoidal
box girders proportioned so as to fit closely within the station platform space. The
type has advantages in using components entirely of steel, which are bolted together
at site and commissioned during temporary possession of existing tracks. Where sufficient depth is available then deck construction is preferable and more economic,
with either twin plate girders or a box girder beneath each rail track.
Simply-supported spans are widely used for railway bridges because:
(a) individual spans can be erected or replaced quickly during temporary track
possession,
(b) uplift is more likely to occur if spans are unequal under heavy railway loading,
(c) fatigue is potentially less critical. Fatigue still tends to govern the design of steel
elements having spans less than about 24 m. Thus cross girders, railbearers and
other short-span members need to use lower working stresses.
4.3.6.3 Pedestrian bridges (Fig. 4.12)
A minimum clear deck width of 1.8 m is usual, increased to 3.0 m or 4.0 m in busy
areas or if a cycleway is also present. Steel provides an efficient solution because
Steel Designers' Manual - 6th Edition (2003)
Selection of type
141
composite
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
cross girder
structure
gauge
1755
typ
7890 typ
steel floor
plate integral
with cross girder
(b)
610
10
II[600
250
350 mm
to
5400
-
-
5400
(c)
Fig. 4.11 Railway bridges. (a) Half-through plate girders; (b) half-through box girders; (c)
composite deck type (spans > 30 m)
Steel Designers' Manual - 6th Edition (2003)
142
Bridges
1800 mm
span
range
remarks
may be erected
complete
)
25
comDosite
(a)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
may be erected
complete
in situ
or PC
units
20
to
50
steel
composite
(b)
use where
construction depth
critical
RHS
)
25.5
(c)
(d)
PG suited over rail
RHSI/ i
through lattice
suited to
members'—
(0
}
lad
CN
interconnect
buildings etc.
50
glazed
1
LI
IJ
(e)
(1)
Fig. 4.12 Pedestrian bridges. (a) Twin universal beam; (b) box; (c) Vierendeel or Warren
girder; (d) half-through universal beam; (e) half-through plate girder; (f) through
lattice
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Codes of practice
143
the entire cross-section including parapets may be fabricated and erected in one
unit. For this reason multiple spans tend to be simply supported and not continuous. Staircase and sloping or stepped ramp spans are also erected as complete units,
and columns are often of steel.
Half-through cross-sections are often used because the shallow construction
depth is able to provide the shortest lengths of staircase or ramp approaches in
urban areas. Half-through rolled beams, Warren truss or Vierendeel girders are
used with rolled hollow section (square or rectangular) members. Floors are steel
stiffened plate, often surfaced with a factory-applied epoxy non-slip surfacing approximately 5 mm thick. Rigid connections between floor and girders provide for
U-frame stability. Staircase approaches may be either steel stringers supporting
steel plate treads or a central spine box with cantilever treads of steel or precast
concrete. Ramps may be similarly formed, but where the span exceeds about 10 m
then half-through construction tends to be used. Spiral approach ramps can be used,
formed in steel using twin rolled sections curved in plan supporting a composite or
steel plate floor. Where adequate construction depth is available, for example when
a pedestrian bridge is required across a motorway or railway cutting, then deck
cross-sections are appropriate with a composite or steel plate floor. Precast concrete
floor units are also suitable. Primary members may be a single box girder, twin
plate girders or twin rolled beams. A single box girder provides a structure of neat
appearance.
4.4 Codes of practice
BS 54003 is used in the UK for the design of all bridges. Part 3 deals with the design
of steel bridges and if composite construction is used then Part 5 must also be
referred to. The Highways Agency and the railway authorities have their own
particular requirements in the form of technical standards which implement and
sometimes vary the clauses in BS 5400. It is expected that Eurocodes will eventually replace BS 5400.
Both BS 5400 and the Eurocodes are based on the limit state design concept,
which means that the verification is carried out for both the serviceability and
ultimate limit states.
The first steps for the preparation of the rules for the traffic loads for bridges, particularly for road bridges, have been undertaken. For the traffic loads on railway
bridges the harmonized uniform load model UIC 714 has been previously adopted
by the different national railway authorities.
Steel Designers' Manual - 6th Edition (2003)
Bridges
144
4.5 Traffic loading
4.5.1 Highway bridges
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Highway bridges in the UK are currently designed for HA loading (a uniformly distributed loading plus knife edge load applied to each traffic lane) together with HB
(abnormal vehicle) loading for structures carrying main highways. Details are given
in BD 37/01, which is a revision of the loading given in BS 5400: Part 2. HA and HB
loading are deemed to allow for dynamic and impact effects. For footways the
normal loading is 5 kN/m2 reduced to 4 kN/m2 where the highway is also loaded. It
is further reduced for longer loaded lengths, similarly to HA loads.
4.5.2 Railway bridges
The uniform load model UIC 714 used by the European national railway authorities is shown in Fig. 4.13: it is also used in BS 5400: Part 2 and reproduced in standard BD 37/01. Where specified, this loading is multiplied by a factor for bridges on
lines carrying heavier or lighter traffic.
Dynamic factors must be applied as shown in Table 4.6. Dimension L is the length
(m) of the influence line for deflection of the element under consideration.
Other loads arising from railway traffic are:
•
•
centrifugal forces on curved track
nosing – 100 kN force acting transversely at rail level
250
250
250
250 kN
80 kN/m
+
Fig. 4.13
80 kN/m
"
no limitation 10.8 ml 1.6 m 1.6 m 1.6 m 0.8 ml no limitation
-1 -t
—I.
UIC and BS 5400 load model (excluding impact)
Table 4.6 Dynamic factors for railway loading
Dimension L
(m)
3.6
>3.6 to 67
>67
Dynamic factor
Bending moment
Shear
2.00
1.67
2.16
0.73 +
÷ L - 0.2
1.00
1.44
÷L - 0.2
1.00
0.82 +
Steel Designers' Manual - 6th Edition (2003)
Steel grades
•
•
145
traction and braking
derailed vehicles (for overturning consider 80 kN/m over 20 m lengths acting on
the edge of the structure).
Fatigue is important for the design of railway bridges because a higher proportion of regular traffic attains the maximum loading compared with highway bridges,
and the vehicles are constrained to run in the same lateral paths.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
4.6 Other actions
Actions other than traffic loading which may need to be considered in one or more
combinations are shown in Table 4.7. The term ‘action’ is used in Eurocodes to
describe all loads or load effects.
4.7 Steel grades
Steels to BS EN 10025 of grade S355 are usual for bridges as they offer a better
cost-to-strength ratio than grade S275. All parts subject to tensile stress are required
Table 4.7 Summary of actions other than those due to traffic loading
Action
Comments
Dead loadsa
Weight of the structure
Superimposed dead loadsa
Finishes and surfacings
Services
Wind
Transverse, longitudinal and vertical
Consider in presence of live load or otherwise
Temperature
Restraint and movement (e.g. flexure of columns)
Frictional restraint of bearings
Effect of temperature difference
Differential settlement
Foundation movements
Earth pressure
Vertical and horizontal pressures from retained material
Erection effectsa
Strength during construction, e.g. stability of steelwork before
composite floor slab cast
Snow load
May be relevant to moving bridges
Seismic
As may be specified by the national authority
Water flow
Flow against bridge supports
a
These actions must be considered during initial design. For most bridge decks actions other than those
from vertical traffic loading, dead and superimposed dead loads are unlikely to have a fundamental
effect.
Steel Designers' Manual - 6th Edition (2003)
146
Bridges
to achieve a specified notch toughness, depending upon design minimum temperature, stress level and material thickness.
Special consideration may be needed if tensile stresses occur only during erection, for example arising from lateral bending in girders due to wind loading, or
during lifting of components. Judgement is necessary but it is often accepted that
modest values of tensile stress can be permitted in such cases without the need for
specific notch toughness provided that the work is not carried out during very low
temperatures. In all other cases appropriate grades having specified notch toughness are necessary.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Weathering steel
To eliminate the need for painting, weathering grades in BS EN 10155 should be
considered. Although it can be shown that the commuted cost of repainting steel
bridges is not of great significance compared with the initial bridge cost, weathering steel is particularly useful in eliminating maintenance where access is difficult –
over a railway and increasingly so over highways. Weathering steel is not suitable
at or near the coast (i.e. within about 2 km of the sea). The Highways Agency
requires sacrificial thickness to be added to all exposed surfaces for possible
long-term corrosion of 1.5 mm per face in a severe industrial environment, 1 mm
otherwise.
4.8 Overall stability and articulation
It is important to consider overall stability of the bridge and its articulation under
temperature effects (see Fig. 4.14). For simply-supported bridges each span must
transfer longitudinal and transverse loads to the foundations while being able to
accommodate movement with suitable bearings and expansion joints. For continuous spans the deck must be pinned at one support with free bearings elsewhere.
Normally only one bearing within the deck width should be pinned longitudinally
so that each girder is free to articulate under traffic loading independently, unless
the pier consists of a separate column beneath each bearing which gives flexibility.
A number of choices are open to the designer but the system used will affect the
design of bearings, bearing stiffeners, expansion joints and the foundations.
For temperature movements:
Movement range: (12 ¥ 10-6 per °C) ¥ temperature range ¥ length from pinned
bearing.
For typical conditions in Europe, for steel or composite decks, ultimate movement
= ±4.5 mm per 10 m of length from mean temperature.
Steel Designers' Manual - 6th Edition (2003)
Overall stability and articulation
EJ
EJ
EJ
147
EJ
001
h1
H
Li
(a)
alternative
half—joint
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(b)
III'
hold—down
C,)
C
0
U
C,)
1111
.1
C')
iinrv
F
Fr
relative
—— articulation
F—fl
E F=O
(c)
centre of
fixity
a pinhead
' sliding
EJ
F
expansion
joint
horizontal force
on bearings
Fig. 4.14 Overall stability and articulation. (a) Simply supported, (b) cantilever and suspended span, (c) continuous, (d) curved viaducts
Steel Designers' Manual - 6th Edition (2003)
148
Bridges
4.9 Initial design
4.9.1 Suspension bridges
For initial design Fig. 4.15 gives approximate formulae5 for making first estimates
of cable size and bending of the stiffening girders. The most severe condition for
the stiffening girder is with approximately one half of the main span loaded
asymmetrically.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
4.9.2 Cable-stayed bridges
Cable-stayed bridges virtually behave as continuous beams with elastic supports at
the cable anchorage points. Cable lengths are adjusted during construction so that
the effect of cable extension under dead load conditions is cancelled out. Provided
that high tensile material is used for the cables then non-linear effects due to selfweight can be shown to be negligible.6 For initial design the deck may be proportioned as a continuous beam, as shown in Fig. 4.16, to which are added:
(a) deck deflections at the cable positions due to extension.
(b) axial forces due to horizontal component of the cable forces.
Where a single plane of cables is used then an initial design check must be made to
ensure that sufficient torsional rigidity is provided (using single or twin box girders)
to control transverse tilt of the deck at mid-span when traffic loading occupies one
carriageway.
4.9.3 Highway bridges – composite deck construction
4.9.3.1 General
Bridges with composite deck construction (Fig. 4.17) represent a high proportion of
the total number of steel bridges built in Europe since 1960. Generally a floor slab
220 mm to 250 mm thick is used composite with rolled sections or plate girders at
spacings up to 3.5 m and depth between (span/20) and (span/30). For spans exceeding 40 m where adequate construction depth is available then twin girders can offer
advantages with typical depth (span/18) to (span/25), and the floor slab is either
haunched over the girders or supported by subsidiary stringers or cross girders. The
slab thickness is determined by its requirement to resist local bending and punching shear effects from heavy wheel loads and needs to be reinforced in both directions: elastic design charts by Pucher are suitable.7 In the hogging regions of
continuous spans the slab will crack and be ineffective in overall flexure unless it is
Steel Designers' Manual - 6th Edition (2003)
Initial design
149
L
—ø H
=L(1+(J))
wL2
8d
H
/
64 d2 x2\'
T =H1+ L4 )
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
=
Ad
w
16 d2
+-----)
16d(
!
!
24d2
15 L f\5
L2
AL (15 — 40
L
I-
+ 288
=
2
16(5-24)
H
Mmax = 0.161 pL2 (4 Eld2
where is the equivalent modulus of
foundation, approximately 1100 kN/m2
Mmax. occurs when
2a
(a)
t / 4 Eld
= —h-- y-— ii)
L
d
stiffness EL
]IIIIj
TW
(cable + deck)
Mmox
(b)
Fig. 4.15 Suspension bridges, approximate formulae (Pugsley). (a) Neglecting stiffness of
deck; (b) including deck stiffness
r
Steel Designers' Manual - 6th Edition (2003)
I
TMmox
n-
I
I
'1TTT
dead + live
M
cable extension
(live only)
J
axial thrust P
approximate analogy
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(0)
i i I'.
_______
£
Si
max.
deflection
dead + live
cable extension
(live only)
oxiol thrust EP
side
span
approximate onology
£
(b)
A
A
A
A
A
Th
A
A
dead + live
cable extension
(live only)
oxiol thrust
P
(c)
Fig. 4.16 Cable-stayed bridges, initial analysis. (a) Anchored side spans (harp shown); (b)
harp; (c) fan
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Initial design
—4)
E
ilk1
0
c0E
'J
0
I-)
E— 2d
4)
0
0
•'U)
0 r)
Fig. 4.17 Highway bridges – composite deck construction. (a) Multiple universal beam (N = 4); (b) multiple plate girder (N
= 4); (c) twin plate girder, haunched slab (N = 2); (d) twin plate girder and stringer (N = 2); (e) twin plate girder
and cross girder (N = 2); (f) multiple box (N = 6)
Steel Designers' Manual - 6th Edition (2003)
151
Steel Designers' Manual - 6th Edition (2003)
152
Bridges
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
prestressed. The longitudinal reinforcement, however, acts as part of the composite
section. Most composite bridges are designed as ‘unpropped’, i.e. the erected steelwork supports its own weight and the concrete slab (including formwork allowance)
until hardened, with composite action being assumed only for superimposed dead
and live loads. Box girders tend to be used for the long spans. Popular forms include
twin box girders, multiple compact boxes and open top trapezoidal boxes.
Plate girder flanges should be made as wide as possible, consistent with outstand
limitations, to give the best achievable stability during erection and to reduce the
number of intermediate bracings. For practical reasons a desirable minimum width
is about 400 mm. A maximum flange thickness of 65 mm is recommended to avoid
heavy welds.
4.9.3.2 Intermediate supports
Intermediate supports often take the form of reinforced concrete walls, columns or
portals. Steel supports may alternatively be used, and tubular columns are efficient,
especially if filled with concrete and designed compositely. Where fewer columns
are required for multiple girders then integral steel crossheads at the supports are
sometimes used.
4.9.3.3 Bracings
For rolled beam or plate girder bridges, lateral bracings are necessary for stability
during erection and concreting of the slab. The bracings are necessary at all supports and when required in hogging regions of continuous spans. If required by the
designer they may be assumed to contribute to the transverse rigidity of the deck
when carrying out an analysis of the transverse distribution of concentrated live
loads. Generally this is advantageous only for decks wider than 20 m. At the abutments the bracing can be a rolled section trimmer composite with the slab and supporting its free end. Over intermediate supports a channel section can be used
between each pair of girders up to about 1.2 m depth. For deeper girders triangulated bracings are necessary.
Intermediate bracings in hogging regions are typically spaced at about 12¥
(bottom flange width). Where bracing is provided across the full width of the bridge,
i.e. between all girders, it increases transverse stiffness significantly. Because of this
stiffness such bracing will attract high stresses under loading that varies across the
width of the bridge. The effect of this behaviour on the fatigue life of the bracing
needs to be considered. Alternatively bracing should be provided only between
neighbouring pairs of girders as shown in Fig. 4.18. This reduces the transverse stiffness considerably and alleviates the problem of fatigue. Such a structure is also likely
to be easier to erect. If the bridge is curved in plan with girders fabricated in straight
Steel Designers' Manual - 6th Edition (2003)
Initial design
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 4.18
153
Lateral bracing of bridge girders
chords they should be located adjacent to the site splices where torsion is induced.
Bracings may be of a triangulated form or of single channel sections between each
pair of girders where up to 1.2 m depth. Bracings are usually bolted to vertical web
stiffeners in the main girders, which may need to be increased in size to accommodate the bolted connections. Angle sections are commonly used with lapped single
shear connections, which permit tolerance in accommodating camber difference
between adjacent girders.
Plan bracing systems may be required for spans exceeding 55 m for temporary
stability under their own weight and that of the wet concrete; they may be removed
after the floor slab is cast.
4.9.3.4 Locations of splices and change of section
Where rolled beams up to 1.0 m deep are used then for the maximum span range
of about 33 m it is convenient to use a constant section with one splice within each
span located at about 0.1–0.2 ¥ span from the internal supports. The beam size will
be determined by the maximum bending and shear effects at the supports where
the slab is cracked.
For plate or box girders the component lengths for shop fabrication should be
the maximum possible consistent with delivery and site restrictions to reduce the
amount of site assembly. For spans up to about 55 m two splices per span are
generally suitable, located at 0.15–0.25 ¥ span from the internal supports, at which
changes of flanges and web thickness should be made. A minimum number of other
flange or web workshop joints should be made, consistent with plate length availability. The decision whether to introduce thickness changes within a fabricated
length should take account of the cost of butt welds compared with the potential
for material saving. Figure 4.19 indicates a basis for considering this optimization.
4.9.3.5 Curved bridge decks
Bridges which are curved in plan may be formed using straight fabricated girders,
with direction changes introduced at each site splice. Alternatively the steel girders
Steel Designers' Manual - 6th Edition (2003)
154
Bridges
L(m) =
rx 10
where r =
7.85(t1—t2)
cost/rn weld
cost/tonne of steel
t2L Ifti
L = minimum
economic length
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
v\c V'rx -'C
—v-- 'C
45
20
50
0
()8
15
12
L
(m)
25
10
----
5
-0
0
—
-
!
—
—
—
—
=—
——
101215 20 25 30 35 40 45 50 55 60
t (mm)
Fig. 4.19 Economy of flange and web thickness changes. The figure gives an indication of
the minimum length (L) for which a selected thickness change will be economic
for flanges and webs of girders. Below this length it will be more economic to continue the thicker plate (t1)
can also be truly curved in plan, in which case the secondary stresses which arise
from torsion must be evaluated. Truly curved girders are appropriate where the
radius of plan curvature is less than about 500 m. It is likely that additional transverse bracings will be needed to reduce these stresses.
Steel Designers' Manual - 6th Edition (2003)
Initial design
155
4.9.3.6 Initial sizes – composite plate girders
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
For economic design an analysis should be carried out of the transverse distribution of concentrated live loads between the main girders of the cross-section. The
floor slab together with any continuous transverse bracing will significantly redistribute the maximum load applied to the most severely loaded girders. Prior to this
the designer will wish to select initial sizes and make an estimate of the total weight
of structural steel. Figures 4.20–4.248 provide initial estimates of flange area, web
thickness and overall unit weight of steelwork (kg/m2) for typical composite bridge
cross-sections as shown in Fig. 4.17.
The figures were derived from approximate designs using simplifying assumptions
2.0
1.9
1.8
1.7
1.6
1.5
1.4
1.3
0
1.1
1.0
0.9
0.8
0.7
0.6
0.5
0,4
1
2
3
girders & slab
4
5
haunch
girder spacing, s
Girder spacing factors
7
stringer
slab
Fig. 4.20
6
8m
girders
1cross
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
U, . p)
('-4
.— C'
'0
0
U,
0
U
,
U')
C'
U')
C'
U,
(-'4
E
C'
('-4
Flange and web sizes – simply-supported bridges
156
-J
C
Fig. 4.21
Steel Designers' Manual - 6th Edition (2003)
Bridges
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
0 O ) N (C
c.J
It)
) CN
0 0(0
u)
E
c.'J
Flange and web sizes – continuous bridges, pier girders
I0
t)
U)
0
U)
0
C
0
0
Fig. 4.22
Steel Designers' Manual - 6th Edition (2003)
Initial design
157
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(N — C'
\IC'__
I 11\j
,
\:t\
I4)
1
I
- I I\
E C
I\'\\
[j Ht 'trH
I+ITIlii
(0 FR
hiI°HVHItTh'
'r)
(N
0
Q
C'
C' C'UC'
jf)IC'f)IC'
lQIUC'IC'L)1C'L,
1C'C'C'ILIC'ILIC'I
I_l '\]
P)
%
't
I
'1II
I
Flange and web sizes – continuous bridges, span girders
158
LI)
If)
—
E
(N
)C'
-jC0
Cl)
Fig. 4.23
Steel Designers' Manual - 6th Edition (2003)
Bridges
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
E
0
U) (0
(0
*(N (N(N
fr (N
* r._)
t) *
I) 0 U)
(N
(N
0000000000
(N
U)
00
0 00U)
0 00
(0
- 0* (N
(N
CE
')
Overall unit weights – plate girder bridges
Initial design
-J
C
0
0
Fig. 4.24
Steel Designers' Manual - 6th Edition (2003)
159
U-)
Steel Designers' Manual - 6th Edition (2003)
160
Bridges
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
for loads and transverse distribution and to achieve correlation with actual UK
bridges.
The sizes indicated do not represent final design, which must be checked by
a form of distribution analysis using the actual specified loads. A two-dimensional
grid analysis is usually employed, with any out-of-plane effects being ignored.
Comparison with three-dimensional analyses has shown that out-of-plane effects
are generally negligible for most composite bridge decks. The following assumptions apply to use of Figs 4.20–4.24.
Deck slab 230 mm average thickness (5.75 kN/m2).
Superimposed dead loads equivalent to 100 mm of finishes (2.40 kN/m2).
Formwork weight 0.50 kN/m2 of slab soffit area.
Steel grade 50 (yield strength 355 N/mm2).
Span-to-girder-depth ratios of 20 and 30.
Webs have vertical stiffeners at approximately 2.0 m centres where such stiffening
is required.
Elastic stress analysis is used for plate girders. If however the plastic modulus is used
for compact cross-sections, then economies are possible.
Steelwork is unpropped during casting of the floor slab.
Sufficient transverse bracings are used such that stresses are not significantly
reduced due to buckling criteria.
Top flanges in sagging regions are dictated by a maximum stress during concreting
allowing for formwork and concreting effects.
Live loading is approximately equivalent to United Kingdom 45HB or HA as
indicated.
Continuous spans are approximately equal in length.
Flange sizes
Figures 4.21–4.23 are applicable to an average girder spacing s of 3.5 m. Figure 4.20
gives a girder spacing factor Kaf, which is multiplied by the flange areas, obtained
above, to give values appropriate to the actual spacing,
e.g. top flange area, Aft = Aft
¥ Kaf
(Figs 4.21–4.23) (Fig. 4.20)
The figures also show actual flange sizes, ranging from 400 ¥ 15 mm to 1000 ¥
75 mm. The flange area of pier girders of continuous unequal spans may be approximately estimated by assuming the greater of the two adjacent spans. End spans of
continuous bridges may be estimated using L = 1.25 ¥ actual span.
Web thickness
Web thicknesses are obtained using Figs 4.21–4.23 applicable to s = 3.5 m.
Adjustment for the actual girder spacing s is obtainable from Fig. 4.20,
Steel Designers' Manual - 6th Edition (2003)
Initial design
161
i.e. web thickness, tw = tw
¥ Ktw
(Figs 4.21–4.23) (Fig. 4.20)
The thickness obtained may be regarded as typical. However, designers may prefer
to opt for thicker webs to reduce the number of web stiffeners. Consideration should
also be given to the use of unstiffened webs of appropriate thickness, i.e. d/t 60–100 depending on shear.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Overall unit weight
Overall unit weight (kg/m2 of gross deck area) is read against the span L from Fig.
4.24 for simply-supported or continuous bridges with L/D ratios of 20 or 30, under
HB or alternatively HA loading and applicable to s = 3.5 m.
Adjustment for average girder spacing s other than 3.5 m is obtainable from Fig.
4.20.
i.e. kg/m2 = kg/m2
¥ Kw
(Fig. 4.24) (Fig. 4.20)
The unit weight provides an approximate first estimate of steelwork weight allowing for all stiffeners, bracings, shear connectors, etc.
For continuous bridges with variable depth, Figs 4.21 and 4.23 may be used to
provide a rough guide, assuming a span-to-depth ratio (L/D) for each span based
upon the average girder depth.
For box-girder bridges a rough estimate may be obtained by replacing each box
girder with an equivalent pair of plate girders.
The mean span for use in Fig. 4.24 should be determined as follows:
+
Ê L41 + L42 + . . . + L4n ˆ
mean span, L = Á
˜
Ë
¯
n
where n is the number of spans.
4.9.3.7 Initial sizes – rolled section beams
An estimate of size for simply-supported spans only may be obtained from Figs 4.25
and 4.2610 for elastic or plastic stress analysis respectively, using universal beams up
to 914 ¥ 419 ¥ 388 kg/m size. For an estimate of the total weight of structural steel
a factor of 1.1 applied to the main beams provides a reasonable allowance for bracings, bearing stiffeners and shear connectors. Figures 4.25 and 4.26 are based upon
a concrete strength of 37.5 N/mm2, and show the required mass per metre of universal beam. Table 4.8 gives reference to the relevant serial size.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
N(N
(N
'4-)
(N
(N
C'-l
(N
0)
—
,-i r)
N)
(N —
r
')
0)
Nwoq(N
(N
NI
sC.J
6uiDods
)C
(N (N NI
NI
Universal beam sizes – simply-supported bridges – elastic design
162
-J
Fig. 4.25
Steel Designers' Manual - 6th Edition (2003)
Bridges
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
\&1 i52 —
S uizods woq
Universal beam sizes – simply-supported bridges – plastic design
0
-J
C
0
0
U,
Fig. 4.26
Steel Designers' Manual - 6th Edition (2003)
Initial design
163
Steel Designers' Manual - 6th Edition (2003)
Bridges
164
Table 4.8 Universal beam sizes
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Reference
Figs 4.25 & 4.26
388
343
289
253
224
201
226
194
176
197
173
147
170
152
140
125
238
179
149
140
125
113
101
Universal beam size
Serial size
Mass per metre
(kg)
914 ¥ 419
388
343
289
253
224
201
226
194
176
197
173
147
170
152
140
125
238
179
149
140
125
113
101
914 ¥ 305
838 ¥ 292
762 ¥ 267
686 ¥ 254
610 ¥ 305
610 ¥ 229
Actual
depth
(mm)
920.5
911.4
926.6
918.5
910.3
903.0
850.9
840.7
834.9
769.6
762.0
753.9
692.9
687.6
683.5
677.9
633.0
617.5
609.6
617.0
611.9
607.3
602.2
References to Chapter 4
1. Johnson R.P. & Buckby R.J. (1986) Composite Structures in Steel and Concrete,
Volume 2: Bridges, 2nd edn. Collins, London.
2. Hayward A.C.G. (1987) Composite pedestrian and cycle bridge at Welham
Green. Steel Construction Today, 1, No. 1, Feb., 5–8. (The journal of the Steel
Construction Institute, UK.)
3. British Standards Institution. Steel, concrete and composite bridges: Parts 1–10.
BS 5400, BSI, London.
BS 5400:
Part
Date
Title
UK Department of
Transport Standard
1
2
3
4
5
6
7
1988
1978
2000
1990
1979
1999
1978
BD15/82
BD37/01
8
1978
9
10
1983
1980
General statement
Specification for loads
Code of practice for design of steel bridges
Code of practice for design of concrete bridges
Code of practice for design of composite bridges
Specification for materials and workmanship, steel
Specification for materials and workmanship,
concrete, reinforcement and prestressing tendons
Recommendations for materials and workmanship,
concrete, reinforcement and prestressing tendons
Bridge bearings
Code of practice for fatigue
BD24/84
BD16/82
BD11/82
–
–
BD10/83
BD9/81
Steel Designers' Manual - 6th Edition (2003)
References
165
4. UIC (1971) Leaflet 702. UIC, 14 rue Jean-Ray F., 75015 Paris, France.
5. Pugsley A. (1968) The Theory of Suspension Bridges, 2nd edn. Edward Arnold,
London.
6. Podolny W. (1976) Construction and Design of Cable-Stayed Bridges. John Wiley
& Sons, New York.
7. Pucher A. (1973) Influence Surfaces of Elastic Plates, 4th edn. Springer-Verlag,
New York.
8. Hayward A.C.G. (2002) Composite Steel Highway Bridges. Corus Construction
Centre, Scunthorpe.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
A worked example follows which is relevant to Chapter 4.
Steel Designers' Manual - 6th Edition (2003)
166
Worked example
Subject
The
Steel Construction
Institute
Chapter ref.
INITIAL DESIGN OF
HIGHWAY BRIDGE
Silwood Park, Ascot, Berks SL5 7QN
4
Made by
Design code
ACGH
Sheet no.
Checked by GWO
BS 5400
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Problem
A composite highway bridge has 3 continuous spans of 24, 40 and
32 m as shown. Overall deck width is 12 m and it carries 45 units
of HB loading. There are 4 plate girders in the cross-section of
1.75 m depth. Estimate the main girder size and the weight of structural steel.
W=12 Pu
D=J.75 LI
I
Span
girder
I
I
Pier
Pier
girder S
girder
Span
girder
I—
24 in
40 in
Span A
Span B
Average girder spacing ‘s’
12
4
=
32 m
Span C
=
3.0 m
Flange and web sizes
From Figure 4.20 kaf
kaf
ktw
=
=
=
0.85 (top flange span girders)
0.87 (generally)
0.95
SPAN A
This is an end span so take
L
=
1.25 ¥ 24 m
=
therefore L/D
=
30 m
1.75 m
so assume L/D
=
20
30 m
=
17
i
4
1
Steel Designers' Manual - 6th Edition (2003)
Worked example
Subject
The
Steel Construction
Institute
Chapter ref.
INITIAL DESIGN OF
HIGHWAY BRIDGE
Silwood Park, Ascot, Berks SL5 7QN
Made by
Design code
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Bottom Flange, Afb
Web, tw
ACGH
Sheet no.
2
Checked by GWO
BS 5400
Top Flange, Aft
4
=
Af (from Figure 4.23) ¥ Kaf
=
0.006 ¥ 0.85
=
Af (from Figure 4.23) ¥ Kaf
=
0.014 ¥ 0.87
=
tw (from Figure 4.23) ¥ Ktw
=
10 ¥ 0.95
=
=
=
0.0051 m2
0.012 m2
9.5 mm
400 ¥ 15
500 ¥ 25
10 mm web
SPAN B
Span girder
L/D
=
40 m
1.75 m
Top Flange, Aft
Bottom Flange, Afb
Web, tw
=
22.9
=
Af (from Figure 4.23) ¥ Kaf
=
0.009 ¥ 0.85
=
Af (from Figure 4.23) ¥ Kaf
=
0.020 ¥ 0.87
=
tw (see Figure 4.23) ¥ Ktw
=
10 ¥ 0.95
This is an end span so take
=
1.25 ¥ 32 m
Therefore sizes as Span B.
=
=
=
SPAN C
L
=
40 m
0.0077 m2
400 ¥ 20
0.0174 m2
500 ¥ 35
9.5 mm
10 mm web
167
Steel Designers' Manual - 6th Edition (2003)
Worked example
168
Subject
The
Steel Construction
Institute
Chapter ref.
INITIAL DESIGN OF
HIGHWAY BRIDGE
Silwood Park, Ascot, Berks SL5 7QN
Made by
Design code
4
ACGH
Sheet no.
3
Checked by GWO
BS 5400
Pier girders
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Take L as the greater of the two adjacent spans
i.e. assume L = 40 m at both supports.
Therefore L/D
=
40/1.75
Top Flange, Aft
=
Af (see Figure 4.22) ¥ Kaf
=
0.015 ¥ 0.87
=
Af (see Figure 4.22) ¥ Kaf
Bottom Flange, Afb
=
=
0.0131 m2
=
= 0.030 ¥ 0.87
Web, tw
22.9
400 ¥ 35
0.026 m2
=
500 ¥ 55
tw (see Figure 4.22) ¥ Ktw
= 16.5 ¥ 0.95
=
15.7 mm
=
30 m
18 mm web
Steel Weight
Span A:
L
=
1.25 ¥ 24 m
Span B:
L
=
40 m
Span C:
L
=
1.25 ¥ 32 m
4
Mean Span
=
4
2
=
4
n
1
4
40 m
1
4
4
4
ÈL + L ... L ˘
È 30 + 40 + 40 ˘ 4
=
ÍÎ
˙˚
ÍÎ
˙˚
3
n
=
37.5
L/D
=
37.5/1.75
kg/m2
=
kg/m2 (from Figure 4.24) ¥ Kw (from Figure 4.20)
=
142 kg/m2 ¥ 1.03
=
=
21
146 kg/m 2
Therefore steel weight
=
146 kg / m2
¥ (24 m + 40 m + 32 m) ¥ 12 m wide =
1000
168 tonnes
Steel Designers' Manual - 6th Edition (2003)
Chapter 5
Other structural applications
of steel
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Edited by IAN DUNCAN with contributions from MICHAEL GREEN,
ERIC HINDHAUGH, IAN LIDDELL, GERARD PARKE,
JOHN TYRRELL and MATTHEW LOVELL
5.1 Towers and masts
5.1.1 Introduction
Self-supporting and guyed towers have a wide variety of uses, from broadcasting
of television and radio, telecommunications for telephone and data transmission
to overhead power lines, industrial structures, such as chimneys and flares, and
miscellaneous support towers for water supply, observation or lighting. These structures range from minor lighting structures, where collapse might have almost no
further consequences, to major telecommunications links passing thousands of
telephone calls or flare structures on which the safety of major chemical plant can
depend. The term ‘mast’ describes a tower which depends for its stability on cable
guys.
5.1.2 Structural types
Steel towers can be constructed in a number of ways but the most efficient use
of material is achieved by using an open steel lattice. Typical arrangements for
microwave radio and transmission towers are shown in Fig. 5.1. The use of an open
lattice avoids presenting the full width of structure to the wind but enables the
construction of extremely lightweight and stiff structures. Most power transmission,
telecommunication and broadcasting structures fall into this class.
Lattice towers are typically square or triangular and have low redundancy. The
legs are braced by the main bracings: both of these are often propped by additional
secondary bracing to reduce the effective buckling lengths. The most common forms
of main bracing are shown in Fig. 5.2.
Lattice towers for most purposes are made of bolted angles. Tubular legs and
bracings can be economic, especially when the stresses are low enough to allow
relatively simple connections. Towers with tubular members may be less than half
the weight of angle towers because of the reduced wind load on circular sections.
169
170
Other structural applications of steel
Fig. 5.1
Lattice towers: (a) microwave tower. (b), (c) and (d) transmission towers
a-
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
However, the extra cost of the tube and the more complicated connection details
can exceed the saving of steel weight and foundations.
Connections are usually arranged to allow site bolting and erection of relatively
small components. Angles can be cut to length and bolt holes punched by machines
Steel Designers' Manual - 6th Edition (2003)
Towers and masts
K
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 5.2
'X'
171
2'
Main bracing arrangements
as part of the same operation. Where heavy-lift cranes are available much larger
segments of a tower can be erected but often even these are site bolted together.
Guyed towers provide height at a much lower material cost than self-supporting
towers due to the efficient use of high-strength steel in the guys. Guyed towers are
normally guyed in three directions over an anchor radius of typically –23 of the tower
height and have a triangular lattice section for the central mast. Tubular masts are
also used, especially where icing is very heavy and lattice sections would ice up fully.
A typical example of a guyed tower is shown in Fig. 5.3.
The range of structural forms is wide and varied. Other examples are illustrated
in Figs 5.4 and 5.5. Figure 5.4 is a modular tower arrangement capable of extension
for an increased number of antennas. The arrangement shown in Fig. 5.5 is adopted
for supporting flare risers where maintenance of the flare tip is carried out at ground
level.
A significant influence on the economics of tower construction is the method of
erection, which should be carefully considered at the design stage.
5.1.3 Environmental loading
The primary environmental loads on tower structures are usually due to wind and
ice, sometimes in combination. Earthquakes can be important in some parts of the
world for structures of high mass, such as water towers. Loading from climatic temperature variations is not normally significant but solar radiation may induce local
stresses or cause significant deflections, and temperatures can influence the choice
of ancillary materials.
Most wind codes use a simple quasi-static method of assessing the wind loads,
which has some limitations for calculating the along-wind responses but is adequate
for the majority of structures. Tower structures with aerodynamically solid sections
and some individual members can be subject to aeroelastic wind forces caused by
vortex shedding, galloping, flutter and a variety of other mechanisms which are
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
172
Other structural applications of steel
Fig. 5.3
Guyed tower
Fig. 5.4
Modular tower
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Towers and masts
Fig. 5.5
173
Flare tower
either poorly covered or ignored by current codes of practice. Such factors have
been responsible for more tower collapses and serviceability failures worldwide
than any shortfall in resistance to along-wind loads.
Most national and international design codes now specify wind loads in terms of
design wind speeds, either mean hourly or gust, that will recur on average once in
a 50 year period (i.e. with an annual probability of 2%). Guidance is sometimes
given on wind shape factors for typical sections and lattices. Consideration of
dynamic response to the wind is not always covered in depth and there is still a
mixture of limit state and working stress codes.
BS 81001 is a recent code in a limit state format specifically written for towers.
Wind loads are specified in terms of a ‘50 year return’ mean hourly wind pressure
together with gust factors which convert the forces to an equivalent static gust. The
overall wind forces calculated using BS 8100 are substantially similar to those
that would be obtained using earlier codes such as CP32 but forces near the tops of
towers are relatively higher due to an allowance for dynamic response. The code
also gives guidance on means of allowing for the importance of particular structures
by adjusting the partial factor on the design wind speed.
Guidance is limited for structures that have a significant dynamic response at their
natural frequencies, and gust factors for guyed towers are specifically excluded from
the scope of the code. Part 4 of BS 8100 is intended to address these aspects.
Steel Designers' Manual - 6th Edition (2003)
174
Other structural applications of steel
The influence of height and topography on wind speed can be significant; this
is covered in some detail in both codes. Ice loads and types of ice are also covered
but neither mentions the very significant influence of topography on the formation
of ice. This has not yet been subject to systematic study but some hill sites are
known to be subject to icing well in excess of the code requirements. The combination of wind and ice loads is even less well understood although some guidance
is given.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
5.1.4 Analysis
In the analysis of towers the largest uncertainty is accurate knowledge of the wind
loads. Highly sophisticated methods of analysis cannot improve this. A static linear
three-dimensional structural analysis is sufficient for almost all lattice tower
structures.
For transmission towers, line break conditions can also be critical. Line breakage
will in general induce dynamic loads in addition to any residual static loads. Detailed
consideration of transmission tower loading is outside the scope of this section.
For lattice towers with large complicated panel bracing, the secondary bracing
forces can be significantly altered by non-linear effects caused by curvature of the
panels under the influence of the design loads. Generally the rules in the codes are
sufficient, but where structures are of particular importance or where there is much
repetition of a design, a non-linear analysis may be necessary.
Dynamic analyses of self-supporting lattice towers are rarely necessary unless
there are special circumstances such as high masses at the top, use as a viewing
platform, or circular or almost solid sections of mast which could be responsive to
vortex shedding or galloping. Knowledge of the dynamic response is also necessary
for assessment of fatigue of joints if this is significant.
For guyed towers the non-linear behaviour of the guys is a primary influence and
cannot be ignored. The choice of initial tension, for example. can have a very great
effect on the deflections (and dynamic behaviour). The effects of the axial loads in
the mast on column stiffness can be significant. Methods of static analysis are given
in the main international codes for the design of guyed towers. Guyed towers can
also be particularly sensitive to dynamic wind effects especially those with cylindrical or solid sections.2
General guidance on the dynamic responses and aerodynamic instabilities of
towers can be obtained from References 3, 4, 5 and 6.
5.1.5 Serviceability
Serviceability requirements vary greatly depending on the purpose of the structure
and its location.
Steel Designers' Manual - 6th Edition (2003)
Towers and masts
175
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel towers and connections arc normally galvanized and are also painted with
a durable paint system if the environment is likely to be polluted or otherwise
corrosive. It is important that regular maintenance is carried out; climbing access
is normally provided for inspections.
Deflections of towers are generally significant only if they would result in a loss
of serviceability. This can be critical for the design of telecommunication structures
using dish antennas. In the past signal losses due to deflection have often been
assessed on the misunderstanding that the deflections under the design wind storm
would occur sufficiently often to affect the signals. Studies have demonstrated that
short periods of total loss of signal during storms smaller than the design wind storm
have a negligible effect on the reliability of microwave links compared with losses
due to regular atmospheric conditions.
5.1.6 Masts and towers in building structures
Consideration of a masted solution arises from the need to provide a greater flexibility in the plan or layout of the building coupled with its aesthetic value to the
project as a whole. At the same time it offers the opportunity to utilize structural
materials in their most economic and effective tensile condition.The towers or masts
can also provide high-level access for maintenance and plant support for services.
The plan form resulting from a mast structure eliminates the need for either internal support or a deeper structure to accommodate the clear span. By providing span
assistance via suspension systems the overall structural depth is minimized giving a
reduction in the clad area of the building perimeter. The concentration of structural
loads to the mast or towers can also benefit substructure particularly in poor ground
conditions where it is cost effective to limit the extent of substructures (Figs 5.6 and
5.7). However, differential settlement can have a significant effect on the structure
by relaxing ties on suspension systems. The consequent load redistribution must be
considered.
Most tension structure building forms consist of either central support or perimeter support, or a mixture of the two. Any other solutions are invariably a variation
on a theme. Plan form tends to be either linear or a series of repetitive squares.
The forces and loads experienced by towers and masts are illustrated in Figs 5.8
and 5.9. In all cases it is advantageous but not essential that forces are balanced
about the mast. Out-of-balance loads will obviously generate variable horizontal
and vertical forces, which require resolution in the assessment of suitable structural
sections.
Suspension ties must be designed to resist not only tension but also the effects of
vibration, ice build-up and catenary sag. Ties induce additional compressive forces
in the members they assist. These forces require careful consideration, often necessitating additional restraints in the roof plane in either the open sections or top
chord of any truss.
Longitudinal stability is created either by twinning the masts and creating a vertical truss or by cross bracing preferably to ground (Fig. 5.10).
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
176
Other structural applications of steel
Fig. 5.6
Traditional long-span structure
Fig. 5.7
Suspension structure
If outriggers are used, as is the case with the majority of masted structures, then
the lateral stability of the outrigger can be resolved in a similar form to the masts
by a stiff truss or Vierendeel section in plan or by plan or diagonal bracing at the
extremity (Fig. 5.11).
5.2 Space frames
5.2.1 Introduction
Steel skeletal space frames are three-dimensional structures capable of very large
column-free spans. These structures, constructed from either individual elements or
prefabricated modules, process a high strength-to-weight ratio and stiffness. Steel
space frames may be used efficiently to form roofs, walls and floors for projects such
as shopping arcades, but their real supremacy is in providing roof cover for sports
stadia, exhibition halls, aircraft hangars and similar major structures.
Steel Designers' Manual - 6th Edition (2003)
Space frames
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
outrigger braced
uplift resistance
(kentledge mass or
tension pile)
/N/
(a)
Fig. 5.8
(b)
Types of perimeter support
\//
Fig. 5.9
c/\7\/
Central support
high level cross bracing
vertical truss
portal action to ground
using perimeter trusses
Fig. 5.10
Longitudinal stability
177
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
178
Fig. 5.11
Other structural applications of steel
(a)
(b)
(c)
Outrigger stability: (a) Vierendeel, (b) plan bracing, (c) diagonal bracing
Steel Designers' Manual - 6th Edition (2003)
Space frames
179
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
5.2.2 Structural types
Space frames are classified as single-, double- or multi-layered structures, which
may be flat, resulting in grid structures, or may be curved in one or two directions,
forming barrel vaults and dome structures. Grid structures can be further categorized into lattice and space grids in which the members may run in two, three or
four principal directions. In double-layer lattice grids the top and bottom grids are
identical, with the top layer positioned directly over the bottom layer. Double-layer
space grids are usually formed from pyramidal units with triangular or square bases
resulting in either identical parallel top and bottom grids offset horizontally to
each other, or parallel top and bottom grids each with a different configuration
interconnected at the node points by inclined web members to form a regular stable
structure.
Single-layer grids are primarily subject to flexural moments, whereas the members
in double- and triple-layer grids are almost entirely subject to axial tensile or compressive forces. These characteristics of single-, double- and triple-layer grids
determine to a very large extent their structural performance. Single-layer grids,
developing high flexural stresses, are suitable for clear spans up to 15 m while
double-layer grids have proved to be economical for clear spans in excess of 100 m.
The main types of double-layer grids in common use are shown in Fig. 5.12.
Skeletal space frames curved in one direction forming single- or double-layer
barrel vaults also provide elegant structures capable of covering large clear spans.
Single-layer vaults are suitable for column-free spans of up to 40 m, which may be
substantially increased by incorporating selected areas of double-layer structure
forming stiffening rings. Double-layer barrel vaults are normally capable of clear
spans in excess of 120 m. Figure 5.13 shows the main types of bracing used for singlelayer barrel vaults.
Dome structures present a particularly efficient and graceful way of providing
cover to large areas. Single-layer steel domes have been constructed from tubular
members with spans in excess of 50 m while double-layer dome structures have been
constructed with clear spans slightly greater than 200 m. Skeletal dome structures
can be classified into several categories depending on the orientation and position
of the principal members. The four most popular types usually constructed in steel
are ribbed domes, Schwedler domes, three-way grid domes and parallel lamella
domes.
Ribbed domes, as the name suggests, are formed from a number of identical rib
members, which follow the meridian line of the dome and span from the foundations up to the top of the structure. The individual rib members may be of tubular
lattice construction and are usually interconnected at the crown of the dome using
a small diameter ring beam.
The Schwedler dome is also formed from a series of meridional ribs but, unlike
the ribbed dome, these members are interconnected along their length by a series
of horizontal rings. In order to resist unsymmetric loads the structure is braced by
diagonal members positioned on the surface of the dome bisecting each trapezium
formed by the meridional ribs and horizontal rings.
I
I
I
P! A I
'' I
N
A
A
A
Steel Designers' Manual - 6th Edition
3!0I (2003)
180
V
Other
/\/\/\ structural applications of steel
viv
I
viv
viv
viv
AAAAAAAAAAAA4
Ivivivivivivivivivivivivivivl
viv
IDuo6o!p
Iv
P!
AOM—oMI
',
,"
\_/
/'
,
/yyyyyy\
ods
ionbs—sjo—
P!6
eDods
P!
E
<
x
xx
.i x i
x x x• x
i
lx
x
x x x x x x x
x x x
x x'
'x'
IxI
x
lx.
/yyyyyy\
/\
-..-
f\
—
A
A
A
A
lxi
IxI
x 'x lx. x
I\
/\
ouo6oip—uo—ionbs
x'
'x lx x
/\
I\
I\
I\
/\
'x'
><><><
><><><
><><
><x
—
—
*
Lattice and space grids
/'
/\
uo—ionbs
Fig. 5.12
'
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
_\
!1°I
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
Space frames
Fig. 5.13
Bracing of single-layer barrel vaults
181
Steel Designers' Manual - 6th Edition (2003)
182
Other structural applications of steel
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Three-way grid domes are formed from three principal sets of members arranged
to form a triangular space lattice. This member topology is ideally suited to both
single-layer and double-layer domes, and numerous beautiful large-span steel threeway domes have been constructed throughout the world.
The steel lamella dome is formed from a number of ‘lozenge’-shaped lamella units
which are interconnected together to form a diamond or rhombus arrangement. The
spectacular Houston Astrodome is an excellent example of this type of construction. This impressive steel double-layer dome was constructed from lamella
units 1.52 m deep and has an outside diameter of 217 m with an overall height of
63.4 m. Figure 5.14 shows the four main dome configurations now in prominent use
worldwide.
5.2.3 Special features
The inherent characteristics of steel skeletal space frames facilitate their ease of
fabrication, transportation and erection on site. There are two main groups into
which the majority of space frames may be classified for assembly purposes: the
particular structure may be assembled from a number of individual members
connected together by purpose-made nodes or alternatively may be constructed by
joining together modular units which have been accurately fabricated in a factory
before transportation to site.
There are numerous examples of ‘chord and joint’ space frame systems available
for immediate construction. These systems offer full flexibility of member lengths
and intersecting angles required in the construction of skeletal dome structures.
Many jointing systems are available; Figs 5.15 and 5.16 show a typical spherical node
used in the MERO system and a cast steel node used in the NODUS system.
The ‘modular’ systems are usually based on pyramidal units which are prefabricated from channel, angle, circular hollow section or solid bars. The individual
units are designed to nest together to facilitate storage and transportation by
road or sea. Most manufacturers of modular systems hold standard units in stock,
which greatly enhances the speed of erection. Figure 5.17 shows typical details of
the prefabricated steel modular inverted pyramidal units used in the Space Deck
System.
Steel space frames are generally erected rapidly without the use of falsework.
Double-layer grids of substantial span can be constructed entirely at ground level
including services and cladding and subsequently lifted or jacked up into the final
position. Dome structures can be assembled from the top downwards using a central
climbing column or tower. A novel approach adopted for the erection of a dome
with a major axis of 110 m and a minor axis of 70 m involved fabrication of the dome
on the ground in five sections, which were temporarily pinned to each other. The
central section was then lifted and the remaining segments of the dome locked into
position as shown in Fig. 5.18.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Space frames
ribbed dome
three-way grid dome
Fig. 5.14
183
Schwedler dome
amelia dome
Dome configurations
5.2.4 Analysis
The analysis of space frames results in the production of large sets of linear simultaneous equations which must inevitably require the use of a computer for their
solution. For these equations to be formulated it is necessary to input into the
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
184
Other structural applications of steel
Fig. 5.15
MERO node connector
Fig. 5.16
NODUS node connector
computer significant amounts of information relating to the topology of the structure and properties of the individual members. This operation can be very timeconsuming unless modern pre-processing techniques which allow rapid data
generation are adopted.7
The members forming double-layer space frames are principally stressed in
either axial tension or axial compression. Consequently it is usual in the analysis
of double-layer space trusses to assume that the members in the structure are
Steel Designers' Manual - 6th Edition (2003)
Space frames
185
unit diagonals all
BS EN 10025 grd 43A
heavy shear 38 dia solid bar
medium shear 28 dia solid bar
light 27 old tube x 3,2 thick
top chord connections
all angle interlaces
made with M12 grd
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
8.8 bolts
boss stud
main tie bar with threaded ends
secondary
tie bar
boss forging
Fig. 5.17
Exploded view of 2.4 m square section of Space Deck
assembly on the ground, with temporarily pinned joints
lift up
Fig. 5.18
Novel method of dome erection
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
186
Other structural applications of steel
pin-jointed, irrespective of the actual joint rigidity. This assumption may lead to the
overestimation of node deflections, but because only three degrees of freedom are
permitted at each node this approach minimizes computer storage requirements and
processing time.
Single-layer grid and barrel vault structures carry the imposed load by flexure of
the members so it is important to include in the analysis the flexural, torsional and
shear rigidity of the members. Linear analysis only is required for the majority of
skeletal space structures but to ensure stability of shallow domes it is essential to
undertake a non-linear analysis of these structures. Large-span space structures may
benefit from a full collapse analysis where the ultimate load-carrying capacity and
collapse behaviour of the structure are determined.8
The behaviour of both single-layer and double-layer space trusses is influenced
to a great extent by the support positions of the structure. The effects of joint and
overall frame rigidity also have a commanding influence and affect the buckling
behaviour of the compression members within the structure. Great care must be
taken in assessing the effective lengths of compression members and it is unlikely
that internal and edge compression members will exhibit similar critical buckling
loads.
5.3 Cable structures
5.3.1 Range of applications
5.3.1.1 Introduction
Most structural elements are able to carry bending forces as well as tension and
compression, and are hence able to withstand reversals in the direction of loading.
Tension elements are unique in that they can carry only tension. In compressive or
bending elements, the loading capacity is often reduced by buckling effects, while
tension elements can work up to the full tensile stress of the material. Consequently
full advantage can be taken of high-strength materials to create light, efficient and
cost-effective long-span structures.
To create useful spanning or space-enclosing structures the tension elements have
to work in conjunction with compression elements. From an architectural point
of view, the separation of the tension, compression and bending elements leads to
a visual expression of the way the structure carries the loads, or at least one set of
loads. Tension structures come in a wide range of forms, which can be broadly categorized as follows:
(1) two-dimensional – suspension bridges
– draped cables
– cable-stayed beams
– cable trusses
Steel Designers' Manual - 6th Edition (2003)
Cable structures
187
(2) three-dimensional – cable truss systems
(3) surface-stressed – pneumatically-stressed
– prestressed.
5.3.1.2 Structural forms
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Suspension bridges (draped cable)
A suspension bridge (Fig. 5.19) is essentially a catenary cable prestressed by
dead weight only. Early suspension bridges with flexible decks suffered from large
deflections and sometimes from unstable oscillation under wind. A system of
inclined hangers proposed originally to reduce deflection under live load has been
employed recently also to counteract wind effects. The suspension cable is taken
over support towers to ground anchors. The stiffened deck is supported primarily
by the vertical or inclined hangers. The system is ideally suited to resisting uniform
downward loads. The principle has been used for buildings, mostly as draped cable
structures.
Cable-stayed beams
Cables assist the deck beam by supporting its self-weight. Compression is taken in
the deck beam so that ground anchors are not required (Fig. 5.20). The cable-stayed
principle has recently been developed for single-storey buildings (see section 5.1 of
this chapter). The cable system is designed for and primarily resists gravity loads; in
buildings with lightweight roof construction the uplift forces, which are of similar
magnitude, are resisted by bending of the stiffening girders. The system is suitable
for spans of 30–90 m and has recently been widely used for industrial and sports
buildings.
Cable truss
The hanging cable resists downloads and the hogging cable resists upload (Fig. 5.21).
If diagonal bracing is used, non-uniform load can be resisted without large deflections but with larger fluctuation of force in the cables.
Three-dimensional cable truss
The classic form of this structure is the bicycle wheel roof, in which a circular ring
beam is braced against buckling by a radial cable system. These radial cables are
divided into an upper and lower set, providing support to the central hub (Fig. 5.22).
Suspension bridge
7, N
Cable-stayed beam
H.
Fig. 5.20
188
Fig. 5.19
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
Other structural applications of steel
Fig. 5.21
Cable truss
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
Cable structures
189
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
190
Fig. 5.22
Other structural applications of steel
Bicycle wheel roof
The system is suitable for spans of 20–60 m diameter. This system has recently been
developed (by David Geiger) into a cable dome, having two or three rings of masts
(Fig. 5.23). The radial forces at the bases of the masts are resisted by circumferential cables. The masts are also cross-cabled circumferentially to maintain their stability. These structures can span up to 200 m.
Surface-stressed structures
A cable network can be arranged to have a doubly-curved surface either by giving
it a boundary geometry which is out-of-plane or by inflating it with air pressure. The
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Cable structures
Fig. 5.23
191
Cable dome
cable net must be prestressed either by tensioning the cables to the boundary points
or by the inflation pressure. The effect of the double curvature and the prestress
is to stiffen the structure to prevent undue deflection and oscillation under loads.
Cable net structures can create dramatic wide-span roofs very economically. They
can be clad with fabric, transparent foil, metal decking or timber boarding, insulation and tiles.
Low-profile air-supported roofs can provide the most economical structure for
covering very large areas (10–50 acres). In designing these structures the aerodynamic profile must be taken into consideration, as must snow loading and the
methods of installation and maintenance.
Steel Designers' Manual - 6th Edition (2003)
192
Other structural applications of steel
5.3.2 Special features
5.3.2.1 Elementary cable mathematics
Load–extension relationship
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Extension
e=
TL
AE
where T = load in cable
L = length of cable
A = cross-sectional area
E = Young’s modulus.
Typical values for materials are given in Table 5.1.
The E value for wire rope applies after the construction stretch has been pulled
out of wire rope by load cycling to 50% of the ultimate tensile strength. In wire rope
the construction stretch can be as much as 0.5%. This is of the same order of magnitude as the elastic stretch in the cable at maximum working load.
Table 5.1 Material properties
Material
Solid steel
Strand
Wire rope
Polyester fibres
Aramid fibres
E
(kN/mm2)
Ultimate
tensile
strength
(N/mm2)
210.0
150.0
112.0
7.5
112.0
400–2000
2000
2000
910
2800
Circular arc loaded radially (Fig. 5.24)
Tension
T = PR
where P = load/unit length radial to cable
R = radius of cable.
Radius of circular arc, R (Fig. 5.25)
Radius
R=
S2 d
+
8d 2
Steel Designers' Manual - 6th Edition (2003)
Cable structures
193
where S = span
d = dip.
Catenary loaded vertically (Fig. 5.26)
Horizontal force
H=
WS 2
8d
Vertical force
V=
WS
2
Maximum tension
T = (H 2 + V 2 ) 2
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
1
where S = span
d = dip
W = vertical load/unit length.
These formulae permit initial estimates of forces in cables to be made. For full and
accurate analysis it is necessary to use a non-linear computer analysis which takes
into account the change of curvature caused by stretch. For well-curved cables the
radius R
P load/rn
Fig. 5.24
Circular arc loaded radially
S
2
Fig. 5.25
Radius of circular arc
S
2
Steel Designers' Manual - 6th Edition (2003)
194
Other structural applications of steel
S
1'
V
W = load/rn
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 5.26
Catenary loaded vertically
hand analysis is accurate enough and gives a useful guide to the forces involved and
hence the sizes of cables and fittings.
Prestressed cable
The straight cable (or flat fabric) is a special problem. To be straight, the cable must
have an initial or prestress tension and theoretically zero weight. In order to carry
load the cable must stretch and sag to a radius R.
For
span
load/unit length
stiffness
pre-tension
tension under load
=S
=W
= EA
= T0
=T
Equilibrium equation T = RW
New length
Strain
Tension
(5.1)
-1
L = 2R sin (S/2R)
= (L - S)/S
T = T0 + EA (L - S)/S
(5.2)
(5.3)
(5.4)
Eliminating R between Equations (5.1) and (5.2), substituting for L in Equation
(5.4) and rearranging gives:
T - T0
2T
SW
=
sin -1
-1
EA
SW
2T
(5.5)
This equation can be solved iteratively for T. The deflection can be found from the
earlier formulae for T and R for a circular arc. It should be noted that:
Steel Designers' Manual - 6th Edition (2003)
Cable structures
195
(1) the extension of a tie with low initial tension is considerably more than TL/EA.
(2) straight cables can be used for load carrying, but either the tension will be very
high or there will be large deflections.
Two-way cable net (Fig. 5.27)
Approximate calculations can be carried out by hand in a similar way to the single
cable calculations above. The basic equilibrium equation for two opposing cables is:
t1
t2
+
=P
R1 R2
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
where P = load/surface area and t1 and t2 are the tensions/unit width, i.e.
t1 =
T1
a1
t2 =
T2
a2
where a1 and a2 are the cable spacings and T1 and T2 are the cable tensions.
The loads on ridge cables and boundaries can be estimated from resolution of the
cable forces if the geometry is known.
The full analysis of cable net structures is a specialized and complicated process
which requires specially written computer programs. The procedure involves the following stages:
T1
T
T2
T2
T2
a,
p
T1
Fig. 5.27
Two-way cable net
Steel Designers' Manual - 6th Edition (2003)
196
Other structural applications of steel
(1) Formfinding: In this stage the cables are treated as constant tension elements
and the geometry is allowed to move into its equilibrium position. On completion of the formfinding stage the cable net geometry under the prestress forces
will be defined.
(2) Load analysis: The prestress model is converted into an elastic model. To do this
the slack lengths, l, of all elements must be set so that
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
t0 = l -
T
EA
The model is then analysed for the defined dead, wind and snow loads.
(3) Cutting pattern definitions: Cable net structures are fully prefabricated with
exact cable lengths so that the prestressed form can be realized. In this stage
the form model will be refined so that the mesh lengths are exactly equal, etc.
The offsets for the boundary clamps must be allowed for. The stressed cable
lengths can then be defined.
During prefabrication of the cables they must be prestretched to eliminate construction struction stretch; they must then be marked at the specified tensions.
Tolerances in fabrication are of the order of 0.02%.
5.3.3 Detailing and construction
5.3.3.1 Cables and fittings
Wire rope cables are spun from high tensile wire. For structural work the cables
are multi-strand, typically 6 ¥ 19 or 6 ¥ 37 with independent wire rope core and
galvanized to Class A. For increased corrosion-resistance, the largest diameter
wire should be used, and cables can be filled with zinc powder in a slow-setting
polyurethane varnish during the spinning process. For even greater corrosionresistance, filled strand or locked-coil strand can be used to which a shrunk-on
polyurethane or polypropylene sleeve can be fitted. Stainless steel, although apparently highly corrosion-resistant, is affected by some aggressive atmospheres if air is
excluded; the resulting corrosion can be more severe than with mild steel.
The simplest and cheapest type of termination is a swaged Talurit Eye made
round a thimble (Fig. 5.28(a)) and connected into a clevis type connection or on to
the pin of a shackle. The neatest and most streamlined fitting is a swaged eye or jaw
end termination (Fig. 5.28(b)). Hot-poured zinc terminations have to be used for
very heavy cables of greater diameter than 50 mm (Fig. 5.28(c)). Epoxy resin with
steel balls can be used as a filler in place of zinc, offering an improvement in fatigue
life at the termination.
On-site connections can be made with bulldog clips but they are ugly and damage
the rope. For cable net construction the standard detail is a three-part forged steel
clamp, of which the two outer parts are identical (see Fig. 5.28(d)). Forging is expen-
Steel Designers' Manual - 6th Edition (2003)
Steel in residential construction
197
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
sive for small numbers and so for smaller structures machined aluminium components may be preferred. Double cables can have a swaged aluminium extrusion
prefixed to each pair of cables, which can then be connected with a single bolt. For
the attachment of net cables to edge cables, forged steel clamps are generally used
(Fig. 5.28(e)). Lower cost alternatives are bent plate or machined aluminium clamps.
Cable life is reduced by corrosion and fatigue. Galvanized cables under cover
suffer very little corrosion; external cables properly protected should have a life of
50 years. Plastic sheathing has the great disadvantage of making inspection of the
cable impossible.
Fatigue investigations have shown that it is wise to limit the maximum tension in
a cable to 40% of its ultimate strength for long-life structures. For structures with
a design life of up to ten years a limit of 50% is acceptable. Flexing of the cables at
clamps or end termination will cause rapid fatigue damage.
5.3.3.2 Rods as tension members
Steel rods are often used as tension members in external situations since they are
stiffer and can be given better protection against corrosion. The rods are usually
threaded at the ends and screwed into special end fittings. In the case of long rods
which can be vibrated by the wind, the end connection must be free to move in two
directions, otherwise fatigue damage will occur.
Since tension members are usually critical components of the structural system,
consideration should be given to using rods or cables in pairs to provide additional
safety.
5.4 Steel in residential construction
5.4.1 Introduction to light steel construction
Light steel framing uses galvanized cold-formed steel sections as the main structural
components. These sections are widely used in the building industry and are part of
a proven technology. Light steel framing extends the range of steel framed options
into residential construction, which has traditionally been in timber and masonry.
The Egan Task Force report called for improved quality, increased use of off-site
manufacture, and reduced waste in construction, and the Egan principles have
been adopted by The Housing Corporation and other major clients in the residential sector. Light steel framing satisfies these Egan principles and it combines the
benefits of a reliable quality controlled product with speed of construction on site
and the ability to create existing structural solutions.
Light steel framing is generally based on the use of standard C or Z shaped
steel sections produced by cold rolling from strip steel. Cold formed sections are
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
198
Fig. 5.28
Other structural applications of steel
Cable fittings: (a) swaged eye, (b) swaged terminator, (c) hot-poured white metal
eye, (d) cross clamp, (e) net and boundary connection
Steel Designers' Manual - 6th Edition (2003)
Steel in residential construction
199
generically different from hot rolled steel sections, such as Universal Beams, which
are used in fabricated steelwork. The steel used in cold formed sections is relatively
thin, typically 0.9–3.2 mm, and is galvanized for corrosion protection.
Cold formed steel sections are widely used in many sectors of construction,
including mezzanine floors, industrial buildings, commercial buildings and hotels,
and are gaining greater acceptance in the residential sector. Light steel framing is
already well established in residential construction in North America, Australia and
Japan.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
5.4.2 Methods of construction
The basic building elements of light steel framing are cold formed sections which
can be prefabricated into panels or modules, or assembled on site using various
methods of connection. The different forms of construction are reviewed in the following sections and are illustrated in Figs 5.29 to 5.31.
Fig. 5.29
Light steel framing using discrete members (‘stick-build’ construction)
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
200
Other structural applications of steel
1
Fig. 5.30
Light steel framing using prefabricated panels
5.4.3 ‘Stick-build’ construction
In this method of construction (illustrated in Fig. 5.32), discrete members are assembled on site to form columns, walls, rafters, beams and bracing to which cladding,
internal lining and other elements are attached. The elements are generally delivered cut to length, with pre-punched holes, but connections are made on site using
self-drilling self-tapping screws, bolts, or other appropriate site techniques.
The main advantages of ‘stick-build’ construction are:
•
•
•
•
•
construction tolerances and modifications can be accommodated on site
connection techniques are relatively simple
manufacturers do not require the workshop facilities associated with panel or
modular construction
large quantities of light steel members can be densely packed and transported in
single loads
components can be easily handled on site.
Steel Designers' Manual - 6th Edition (2003)
Steel in residential construction
201
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
h
fr, t;
I' 1e
Fig. 5.31
Modular construction using light steel framing
‘Stick-build’ construction is generally labour intensive on site compared with the
other methods, but can be useful in complex construction, where prefabrication
is not feasible. This form of construction is widely used in North America and
Australia, where there is an infrastructure of contractors skilled in the technique.
This stems from a craft tradition of timber frame construction that now uses many
power tools. In these countries, traditional timber contractors have changed to light
steel framing with little difficulty.
5.4.4 Panel construction
Wall panels, floor cassettes and roof trusses may be prefabricated in a factory and
later assembled on site, as in Fig. 5.33. For accuracy, panels are manufactured in
purpose-made jigs. Some of the finishing materials may be applied in the factory, to
speed on-site construction. Panels can comprise the steel elements alone or the
facing materials and insulation can be applied in the factory. The panels are connected on site using conventional techniques (bolts or self drilling screws).
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
202
Fig. 5.32
Other structural applications of steel
‘Stick-build’ site construction using wall studs and floor joists
The main advantages of panel or sub-frame construction are:
•
•
•
•
speed of erection of the panels or sub-frames
quality control in production
reduced site labour costs
scope for automation in factory production.
The geometrical accuracy and reliability of the panels and other components is
better than with stick-build construction because panels are prefabricated in a
factory environment. The accurate setting out and installation of foundations is a
key factor to achieve rapid assembly of the panels and to obtain the maximum
efficiency of the construction process.
5.4.5 Modular construction
In modular construction, units are completely prefabricated in the factory and may
be delivered to site with all internal finishes, fixtures and fittings in place, as illustrated in Fig. 5.34. Units may be stacked side by side, or one above the other, to
form the stable finished structure.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
Steel in residential construction
Fig. 5.33
Site assembly of light steel wall panels
Fig. 5.34
Modular construction using light steel framing
203
Steel Designers' Manual - 6th Edition (2003)
204
Other structural applications of steel
Modular construction is most cost-effective where large production runs are possible for the same basic configuration of modular unit. This is because prototyping
and set-up costs, which are essentially independent of scale, can be shared across
many units. Details of the use of modular construction in residential construction
are given in reference 9.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
5.4.6 Platform and ‘balloon’ construction
‘Stick-build’ or panel components may be assembled in either ‘platform’ or ‘balloon’
construction, as illustrated in Fig. 5.35. In platform construction, walls and floors are
built sequentially one level at a time, so the walls are not structurally continuous.
In some forms of construction, loads from the walls above are transferred through
the floor joists to the wall below.
In ‘balloon’ construction, the wall panels are often much larger and are continuous over more than one storey. Such panels are more difficult to erect than
single storey height panels and have to be temporarily braced whilst the floors are
installed. The main advantage of this approach is that loads from the walls above
are transferred directly to those below.
In both forms of construction, the external cladding or finishes are generally
installed and attached to the frames on-site.
5.4.7 Material properties
The galvanized strip steel from which the light steel framing is formed is usually
designated as either grade S280GD or grade S350GD to BS EN 1014710 (formerly
Fe E 280 G or Fe E 350 G). These designations indicate the yield strength (280 or
350 N/mm2) and the fact that the material is galvanized with a minimum G275
coating. Cold formed steel sections are usually rolled from galvanized sheet steel
that is typically 0.9–3.2 mm thick. The normal thickness of zinc coating (275 g/m2)
has excellent durability for internal applications. Heavier coatings are available for
more aggressive external environments.
5.5 Atria
5.5.1 General
An atrium is an enclosed courtyard space often rising through several storeys of a
building. Its function is usually to provide daylighting within the plan of a building.
Often it provides covered circulation space between floor levels which readily
Fig. 5.35
<0<00)
00
0
C)
C
Cl)
0
0
0
C)
0
0)
‘Platform’ and ‘balloon’ forms of panel construction
-+
CD
0.
'0
DC)
0)
<CD
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
0,
<Do)
0) _Q)
m
Steel Designers' Manual - 6th Edition (2003)
Steel Designers' Manual - 6th Edition (2003)
206
Other structural applications of steel
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
orientates the building occupants. In common with all aspects of building design,
there are a number of design considerations which come into play in the creation
of an atrium space. These broad design requirements generate specific technical
requirements for the design of the structure. These need to be taken into consideration at the developmental stages to ensure that an elegant design, consistent
with all of the technical criteria, can be achieved.
The following design issues are common to the design of an atrium space:
•
•
•
•
•
•
•
•
•
•
•
•
Structural form
Type of cladding
Tolerance and deflection
Finishes and detailing
Erection of structure, cladding and fit out
Procurement
Maintenance of the cladding, structure and fit out
Internal and external cleaning
Acoustic performance of the space
Cost
Fire safety of the atrium and surrounding spaces
Provision of heating and cooling to achieve required environmental conditions.
The following sections look at some of the technical implications for the structural
design and specification imposed by these design issues.
5.5.2 Structural aspects
5.5.2.1 Structural form
For a new building there will be significant choice in the types of structural systems
that can be adopted, as the perimeter supports can be designed accordingly. Where
an enclosure is being added to an existing building, available support will depend
on the capabilities of the existing structure and may determine the appropriate
options.
The structures in Figs 5.36 to 5.40 illustrate a series of solutions in which there is
a generally decreasing reliance on the perimeter structure.
As a group the arch, the beam and the dome are potentially the most structurally
efficient solutions but rely totally on the support provided by the perimeter structure. In the case of an existing building, the following should be considered:
•
•
The strength and integrity of the existing structure if this is to be used for support.
The percentage increase in vertical load imposed on the existing structure by the
new. This will be smaller where new loads are introduced towards the lower levels
of existing walls and columns than at the top.
Steel Designers' Manual - 6th Edition (2003)
Atria
207
mechanical extraction or natural
smoke ventilation in wind suction
area
mechanical smoke
extraction in lieu
thrust from arch resolved
into columns and floor,
horizontal forces go into
wall or perimeter roof
of fire — resistant
glazing in elevations
diaphragm
F
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
low— level arch
shear wall
gives overall
fire— resistant
glazing
Fig. 5.36
Arch
simple profile makes addition of
cleaning gantry straightforward
low—level roof
additional columns ore
avoids costs of
additional fire
more in scale for low
safety provisions
at high level
Fig. 5.37
Beam or truss
option
Steel Designers' Manual - 6th Edition (2003)
Other structural applications of steel
208
enclosure 2—3 m above
occupied space for more
bearings or horizontal tronslotion of strut
relieve temperature effects
effective environmental
control
good location
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
for
pl
Fig. 5.38
airflow to offset heat
gains and
theyoc
ere
shading excludes
summer sun but allows
winter sun to penetrate
heating or cooling
Half portal
profile of roof steelwork forms
natural smoke reservoir and
venting at top
shading
perimeter structure provides
nominal lateral restraint to
roof enclosure but majority
of vertical load is supported
by interior columns
fire resistant
base of smoke reservoir 3 m —i
higher than 1st floor deck
allows means of escape
Fig. 5.39
•
existing perimeter
structure with limited
load bearing capacity
Beam and column
The horizontal forces imposed at the supports. Horizontal thrusts introduced
at lower levels will have smaller resultant overturning effect than at a higher
level.
Failure to investigate the implications of the choice of structural form could result
in a costly and clumsy solution.
As town centres and streets are redeveloped to compete with out-of-town shopping, the need for enclosure is a common solution, often considered. In these cases,
the adjacent buildings are typically of a wide variety of constructions of various
heights and condition. The solutions defined in Fig. 5.39 may be appropriate to avoid
the complexities of resolving the various boundary conditions. The perimeter struc-
Steel Designers' Manual - 6th Edition (2003)
Atria
lateral stability provided by portal action;
perimeter accommodation provides wind
shielding; longitudinal stability is provided
by portal action or braced frames
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
water run — off to new system
panel propped top
and bottom but free
vertical and horizontal
dilferential movement
allowed
maisonettes
choice of sprinklers
and other fire
protection measures
excluded from
enclosure to reduce
could affect final form
of the structure
and smoke
Fig. 5.40
209
risk from fire
Portal frame
ture is required to support a small proportion of the vertical load and relatively
small horizontal loads from wind and stabilizing forces for the new columns, which
does not generally pose a problem since there is usually some redundant horizontal capacity due to the wind shielding effect of the new enclosure. The portal style
solution (Fig. 5.40) is substantially self-supporting.
Temperature effects are often a governing design consideration for an atrium
structure due to exposure to heating from sunlight. Temperatures in steelwork
painted black can be 30° higher compared with white where directly exposed to the
sun’s radiation. It is important to allow for movement of the structure, to prevent
locked-in stresses from building up. Releases of this sort are often incompatible with
providing horizontal stability. The choice of structural form needs to reflect these
opposing requirements. For example, a statically determinate structure such as a
three pinned arch can accommodate temperature movements and differential
settlements without generating secondary internal forces (see Fig. 5.41).
5.5.2.2 Type of cladding
Atria generally require transparent or translucent forms of cladding. Glazing is a
traditional solution but is relatively heavy. If the space beneath is heated the thermal
performance may dictate double-glazing, which is both heavy and costly. Other
Steel Designers' Manual - 6th Edition (2003)
210
Other structural applications of steel
joint
horizontal
movement
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
frames with pin joints
allow temperature relief
by simple vertical and
horizontal movement and no
primary forces are induced
(0)
z::
/
centre of
movement c'
direction of lateral restraint
free temperature movement
(b)
Fig. 5.41
Movements caused by temperature changes
Steel Designers' Manual - 6th Edition (2003)
Atria
211
forms of cladding such as ETFE foil cushions can offer a lightweight, thermally
efficient and cost-effective alternative to glazing.
Glazing systems may be framed or unframed. The allowable spans of the systems
vary. The relationship between the layout of the primary structure and the cladding
system will determine whether there is a need for a secondary system of support.
It is important to consider this relationship at concept stage to achieve the minimum
of elegantly arranged members. The following dimensions provide guidance:
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
•
•
•
Planar glazing – spans in either direction around 2 m
Framed systems – spans 4–6 m by 2 m
ETFE foil cushions – spans 6–8 m in either direction.
Manufacturers should be consulted for more precise information.
Framing systems are generally made from aluminium extrusions, and therefore
have thermal expansion and contraction characteristics which differ from steelwork.
This needs to be considered in the detailing of the connectivity between the two
elements.
Tolerance between cladding and supporting steel framework also needs careful
consideration. The accuracy with which steelwork can be fabricated and erected is
generally lower than that of the cladding system. This incompatibility should also
be accounted for in the detailing of connections.
Deflections of the atrium structure under loads can have significant bearing on
the cladding system. Some glazing systems require a minimum slope to ensure their
water tightness. The deflected shape of the supporting structure must satisfy these
requirements under all relevant loading conditions. This may require the structure
to incorporate a precamber from the theoretical final geometry. The cladding system
must also be capable of accommodating any differential movements of the supporting structure.
5.5.2.3 Finishes and detailing
The detailing of individual members and their connections is crucial to achieving a
satisfactory architectural treatment. It is commonly the fabricator’s responsibility to
detail connections, and often this expertise can inform the design process. However,
it is important to ensure that aesthetic considerations are properly defined and controlled. The following should be considered:
•
•
•
Nature of the connection between components of the primary structure
Connection of the cladding system
Connection of sundry components such as access equipment, lighting, etc.
Connections can be expressed or ‘invisible’. The strategy for connecting components
should take into account the practicality of achieving the desired outcome. Site
Steel Designers' Manual - 6th Edition (2003)
212
Other structural applications of steel
welding can offer ‘invisible connections’ within certain limitations. The following
should be considered:
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
•
•
•
•
•
The type of welding technique and position required to form a connection
Butt welding can be ground flush whereas fillet welds cannot
The type of weld testing required to verify connection adequacy
The effects of heat distortion on the structural strength and geometry
The effect of welding on shop-applied finishes.
Where connections are aesthetically critical, sample workmanship should always be
called for as a basis for selecting a contractor. Samples should be kept as a record
against which to judge finished workmanship.
Steelwork will require a paint finish for reasons of aesthetics and durability. Paint
technology provides an ever increasing number of options. However, the choice can
be guided by consideration of the following criteria:
•
•
•
•
•
•
The steelwork fabricator who delivers and erects the steel will inevitably damage
any paint applied at the works no matter how carefully the protection measures
are prepared.
The contractor who has a very tight programme may like to see primer, undercoat and finish coat on the steelwork applied at the works since it may be very
difficult to apply the finish in high-level steelwork without extensive temporary
works.
Very hard paint finishes are more difficult to damage although when the
inevitable occurs they are usually more difficult to repair to a good standard.
If the finish-coat is applied on site, some parts of the steelwork may be difficult
to reach and the standard of finish will not be as good as that achieved in the
paint shop.
Unless touching up to damaged areas is very minor, the result is unlikely to
be satisfactory and therefore a full site applied decorative finish may prove
necessary.
The choice of system will also depend on the fabricator’s paint shop and its relationship with the paint supplier.
5.5.2.4 Erection
The engineer should always consider the erection method for a structure. In the case
of atria this may be of particular concern for the following reasons;
•
•
•
•
The site is constrained by surrounding buildings
The application of load to existing structures to be carefully controlled
The need to precamber the geometry
Restrictions on the location, reach and capacity of craneage
Steel Designers' Manual - 6th Edition (2003)
Atria
•
•
•
213
Limits on the size of components
The formation of neat site connections
Safe access for operatives for the erection of the primary structure, cladding and
other items.
The erection methodology may preclude the use of some structural forms. It is,
therefore, essential to consider this aspect at concept stage.
5.5.2.5 Procurement
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
An atrium structure will have several complex interfaces:
•
•
•
The cladding
Cleaning equipment
Supporting structure
The management of these interfaces is critical to the successful implementation of
the design. If the structure, cladding, etc are procured through separate contracts it
is even more critical to define the technical requirements and responsibilities at
interfaces. The following issues should be addressed:
•
•
•
•
Forces and movements to be accommodated
Compatibility requirements of materials
Design responsibility for components and fixings
Definition of overall theoretical and as-built geometry.
It can be convenient to incorporate several packages of work into a single contract.
This can reduce risk associated with the management of interfaces. This route may
increase cost due to the increased responsibilities carried by the contractor.
5.5.2.6 Maintenance
Ongoing and long term cleaning and maintenance will be required and may impact
on the design. Ongoing work will require a simple safe means of access. It will be
necessary to account for the loading and connection of gantries, walkways, harness
points, etc., in the design of the primary structure.
Long term maintenance could be facilitated by independent access systems such
as scaffolding. The main impact on the design of the structure will be the quality of
the specification. The life to first maintenance of the finishes and components will
increase with initial cost. This needs to be balanced by the costs of the maintenance
activity. The client should be consulted on the options available and choose the most
Steel Designers' Manual - 6th Edition (2003)
214
Other structural applications of steel
appropriate option. This will depend on his/her long term interest in the building
fabric.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
5.5.2.7 Acoustics
Large enclosed spaces may require acoustic treatment to limit reverberation times.
Hard surface finishes can exacerbate the problem. These considerations may affect
the choice of finishes within an atrium or necessitate the provision of acoustic baffles
or other treatments. This is only of particular concern if the atrium structure is
required to support the loads of such components. If so, these must be allowed for
in the design.
ETFE foil cushions are substantially transparent from an acoustic point of view,
which may make them unsuitable for city centre applications. They may also
generate significant noise due to the impact of rain.
5.5.2.8 Cost
There are several factors that affect the cost of steel structures of this type:
•
•
•
•
•
•
Type of section used
Complexity of fabrication
Complexity of installation
Type of finish
Quality of workmanship
Procurement arrangements
It is common to budget steelwork on the basis of rates applied to each tonne of
material. While this is a useful indication of cost, it is important to recognize that
the appropriate rates for these types of structure may be two to four times greater
than those for standard structural steelwork.
5.5.3 Fire engineering
5.5.3.1 Background
The fundamental principle of atrium fire safety design is to ensure that the design
standard is not compromised by the presence of the atrium; the standard of the
building with the atrium should not be less than the standard of an equivalent building without the atrium. This section describes the considerations and processes
Steel Designers' Manual - 6th Edition (2003)
Atria
215
involved in the development of the atrium fire strategy, concluded by summarizing
the implications for the structural design.
The primary objective of the fire strategy is to achieve an acceptable standard of
life safety. BS 5588: Part 7 is the relevant ‘life-safety’ guidance for the incorporation
of atria into buildings, and is applicable in all but limited circumstances. A package
of fire safety provisions is developed to address the following requirements:
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
•
•
•
•
•
Means of escape
Compartmentation
Structural fire resistance
External spread of fire
Facilities for the fire service. Smoke clearance is relevant for this situation.
This package depends on the size and use of the building. As examples:
(1) In a small office building, the design is for simultaneous evacuation and there
is no need for compartmentation between storeys. Thus in this situation the
atrium can be open, as the consequences of a fire anywhere in the building with
an atrium are no different (in terms of life safety) than those in the same building without an atrium.
(2) In tall office buildings, storey-to-storey compartmentation is defined to enable
phased evacuation; this vertical compartmentation needs to be maintained in
the building with an atrium.
(3) In a hospital, as well as vertical compartmentation there is a need for horizontal compartmentation. This enables progressive horizontal evacuation of the fire
compartment, with all other compartments remaining in place.
(4) With an atrium it is necessary to consider thermal radiation across the atrium,
to ensure that this does not create untenable conditions in a non-fire compartment on the same storey adjacent to the atrium.
In situations where occupants remain in the building, the psychological effects of
smoke in the atrium must also be considered, even though there may be no other
risk posed by the fire.
The property protection and business continuity implications must be carefully
considered. Even in situations where there is an acceptable standard of life safety,
the atrium provides the opportunity for substantial spread of smoke, especially
with an unenclosed atrium. These additional requirements are usually specified by
the client, operator, or insurer and not regulated under legislation. However, it
should be noted that there are Local Government Acts (such as Section 20 in
London) and national legislation outside England which have property protection
implications.
The design procedure of the atrium can be generalized by consideration of:
•
•
Use of the accommodation adjacent to the atrium
Use of the atrium base
Steel Designers' Manual - 6th Edition (2003)
216
•
•
Other structural applications of steel
Performance of the atrium façade
Control of smoke within the atrium.
Effective smoke control is usually the key to the success of the atrium fire strategy.
This can be achieved either by preventing smoke entering the atrium (e.g. an atrium
façade with fire-resisting performance and a fire-sterile atrium base) or, alternatively, by managing any smoke within the atrium. The atrium can be divided into
three zones:
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(1) Atrium base
(2) The smoke reservoir at the top of the atrium, where any hot smoke will collect
and can be extracted
(3) Remainder of the atrium.
Different design considerations will apply for these different zones.
The smoke management will normally take advantage of a smoke extract system.
The design of this system depends on the quantity of smoke within the atrium and
the extract capacity. The quantity of smoke is dependent on the fire size and width
of spill, where the smoke spills into the atrium. Complex spill geometries create
large quantities of smoke. Smoke extract can be by either mechanical extract or
natural systems. In either case it is necessary to arrange for low-level make-up for
replacement air. Wind-effects on natural systems must be addressed. The extract can
serve either to limit the depth of the smoke reservoir, or to limit the smoke reservoir temperature, or both. In either case the fire size must be controlled.
The fire size is normally controlled either by use of sprinklers or by management
control. Management control defines islands of fire load, the size of the island
dependent on the amount of combustibles, separated to ensure that the fire does
not spread between islands. Typical rules are 10 m2 islands separated by 4 m.
Sprinklers are usually assumed in smoke control calculations to limit the fire size
to a maximum area equal to the sprinkler-grid area. In practice, sprinklers may
extinguish the fire or control it to a substantially smaller area. As an alternative to
sprinklers or management control, a fire-engineered approach can be used to
account for the likelihood of a particular fire size occurring, given the use of the
accommodation (‘natural fire’ based design).
It is important also to account for cool smoke, which cannot be assumed to rise
(and may even drop). Where the smoke cannot be assumed to be hot enough to rise
to the reservoir by buoyancy, it will be carried by the airflows that exist within the
atrium. CFD analysis may be required to design the cold smoke system. Careful
coordination with the building services engineer will enable the best value solution,
particularly with natural ventilation systems, as the cold smoke is primarily carried
by the environmental flows. An upward flow of air is beneficial for both the fire
strategy and the environmental strategy.
Use of the atrium base can be enabled, provided that the rules for smoke and fire
spread are respected. It is difficult to use sprinklers for controlling the fire size in
the atrium base, as there is no ceiling above it (the atrium roof is usually too high
Steel Designers' Manual - 6th Edition (2003)
Atria
217
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
to enable effective operation of sprinklers), and thus fire load control is either by
management control or by an engineered approach.
Where the atrium facade is required to have a fire performance, this is defined
either to prevent the fire breaking into the atrium, or to prevent the smoke breaking back in. Where an enclosure is required to have a fire-resisting performance, this
can either be achieved by a formal fire-resisting system, or alternatively a smokeretarding construction may be used in combination with an extract system to limit
smoke temperatures. There is currently no standard for smoke-retarding construction, and thus it is often preferable to specify fire-resisting construction to avoid
complicated specifications and enhanced site control.
5.5.3.2 Structural implications
The structural implications of the above can be summarized as follows:
•
•
•
Structural fire resistance: The required fire resistance for elements of structure
is equivalent to that of the building without the atrium.
Facade: The structure needs to ensure that the facade can achieve its required
performance. This may equate to ensuring that the deflections in a fire condition
are limited or, more simply, by specifying a fire-resisting system for the facade.
Although BS 5588: Part 7 defines a performance criterion for ‘smoke retarding’
construction, consideration should be given to specifying fire-resisting construction instead; this is because there is no standard specification for smokeretarding construction, and there will be necessity to develop ad-hoc material
specification, construction methods and QA procedures to ensure that the system
meets the required performance and to demonstrate this to the regulators.
Atrium roof: Whilst structural elements solely supporting the roof neither formally nor usually require fire resistance, it is prudent to consider the effect of
collapse in development of the fire strategy. Thus potentially there may be a
requirement for fire resistance. Also, as normal, where the roof structural elements support elements requiring a fire-resistance (such as the atrium facade),
then these structural elements must achieve the required fire performance.
The fire strategy for a building with an atrium is complex, particularly where there
is a need to incorporate local government Acts or property protection requirements,
or where there are non-standard client, architectural, or engineering objectives.
A fire-engineered approach will enable the optimum strategy and best value to
be achieved. Structural fire engineering can be used with an atrium building in the
same way as for a building without an atrium, taking account of the inherent
performance of the structure to reduce or omit the need for fire protection. Fire
engineering enables best integration of the fire strategy within the design, to ensure
that all systems are tailored to meet the precise performance and reliability requirements (for fire and all other requirements).
Steel Designers' Manual - 6th Edition (2003)
218
Other structural applications of steel
5.5.4 Environmental engineering
5.5.4.1 General
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
There is an intimate relationship between thermal physics, smoke ventilation and
structural form in atria, and the design of the larger structural shape is influenced
by them. In addition, the materials used will have to meet the fire and physics
criteria to perform as required.
Environmentally, an atrium can be used as a buffer zone between indoors and
outdoors. This zone can be exposed to the external environment, providing protection from wind, rain, sun, etc., and/or a controlled environment similar to the indoor
conditions of the building.
Important considerations in atrium design include:
•
•
•
•
•
•
•
•
•
Characteristic height for stack ventilation
Areas of glazing – optimization of natural light to the space beneath
Need for temperature-controlled conditions
Risk of condensation (winter) – heated or unheated
Spatial separation (environmentally) from internal spaces
Ventilation manually openable or BMS controlled
BMS ventilation control – temperature or wind direction sensitive
Shading (internal, external), material options
Smoke reservoirs.
5.5.4.2 Shading
In the atrium, heat transfer by radiation is important, and shading systems may be
required to reduce direct solar radiation and allow diffuse solar radiation for natural
light. The type of shading – internal or external – will usually be related to its
aesthetics, performance and maintenance costs. Shading can enhance stack effect by
absorbing solar heat and contributing to buoyancy driven flows.
5.5.4.3 Ventilation
Stack ventilation is used when cross ventilation is not possible and single sided ventilation cannot provide sufficient air change rates. Strong temperature gradients can
occur in atria, stack effect being enhanced by wind forces in some cases. Temperature differences can be used both during the day and during the night to enhance
night cooling.
The atrium needs to extend above the height of the stack so that the neutral
Steel Designers' Manual - 6th Edition (2003)
References
219
pressure level is high enough to provide sufficient driving force to draw air in and
out of the building.
It may be necessary to add an atrium to an existing building if the shape of the
building is outside the limits for cross ventilation (15 m deep). This allows for natural
ventilation to all areas, improved natural light penetration and beneficial effects on
internal circulation.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
5.5.4.4 Smoke
Smoke control can require a large space at the top of the atrium to act as a smoke
reservoir. This reduces the amount of smoke moving into upper floors and allows
for full evacuation of the building. Extraction of toxic gases can be by means of
mechanical and/or natural vents located at the top of the atrium. Smoke venting via
the atrium can be achieved by taking advantage of the strong buoyancy of smoke,
but the use of channelling screens may be necessary. For instance, automatic window
or curtain closing on the atrium perimeter can be used as a screen mechanism.
References to Chapter 5
1. British Standards Institution (1986) Lattice towers and masts. Part 1: Code of
practice for loading. BS 8100, BSI, London.
2. British Standards Institution (1972) Loading. Chapter V: Part 2: Wind loads.
CP3, BSI, London.
3. Construction Industry Research and Information Association (CIRIA) (1980)
Wind Engineering in the Eighties. CIRIA Conference Report 12/13 Nov.
4. Engineering Sciences Data Unit. Wind engineering sub-series (4 volumes).
ESDU International, London.
5. Vickery B.J. & Basu R.I. (1983) Across wind vibrations of structures of circular
cross section. Journal of Wind Engineering and Industrial Aerodynamics, 12,
49–97.
6. International Association for Shell and Spatial Structures (IASS) (1981) IASS
Recommendations for Guyed Masts. IASS, Madrid.
7. Nooshin H. & Disney P. (1989) Elements of Formian. Proceedings of 4th Intl
Conf. Civ. and Struct. Engng Computing (Ed. by H. Nooshin), pp. 528–32.
University of Surrey, Guildford, UK.
8. Parke G.A.R. (1990) Collapse analysis and design of double-layer grids. In
Studies in Space Structures (Ed. by H. Nooshin), pp. 153–79. Multi-Science
Publishers.
9. Gorgolewski M.T., Grubb P.J. & Lawson R.M. (2001) Modular Construction
using light steel framing: Design of residential buildings. SCI-P302. The Steel
Construction Institute, Ascot, Berks.
Steel Designers' Manual - 6th Edition (2003)
220
Other structural applications of steel
10. British Standards Institution (1992) BS EN 10147: Specification for continuously
hot-dip coated structural steel sheet and strip. BSI, London.
Further reading for Chapter 5
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Section 5.2
Bell A.J. & Ho T.Y. (1984) NODUS spaceframe roof construction in Hong Kong.
Proceedings 3rd International Conference on Space Structures (Ed. by H.
Nooshin), pp. 1010–15. University of Surrey, Guildford, UK.
Bunni U.K., Disney P. & Makowski Z.S. (1980) Multi-Layer Space Frames. Constrado, London.
Makowski Z.S. (1984) Analysis, Design and Construction of Braced Domes.
Granada, St Albans.
Makowski Z.S. (1985) Analysis, Design and Construction of Braced Barrel Vaults.
Elsevier Applied Science Publishers, Barking, Essex.
Parke G.A.R. & Walker H.B. (1984) A limit state design of double-layer grids. Proceedings 3rd International Conference on Space Structures (Ed. by H. Nooshin),
pp. 528–32. University of Surrey, Guildford, UK.
Supple W.J. & Collins I. (1981) Limit state analysis of double-layer grids. Analysis,
Design and Construction of Double-Layer Grids (Ed. by Z.S. Makowski), pp.
93–117. Applied Science Publishers, Barking, Essex.
Section 5.3
Liddell W.I. (1988) Structural fabric and foils. Kerensky Memorial Conference –
Tension Structures, June, Institution of Structural Engineers.
Troitsky M.S. (1988) Cable Stayed Bridges: Theory and Design, 2nd edn. BSP Professional, London.
For surface stressed structures refer to publications of the Institut für Leicht
Flächentragwerke, Universität Stuttgart, Pfaffenwaldring 14 7000, Stuttgart 80
IL 5 Convertible roofs
IL 8 Nets in nature and technics
IL 15 Air hall handbook
Section 5.4
Chung K.F. (1993) Building design using cold formed steel sections: Worked
examples. The Steel Construction Institute, Ascot, Berks.
Clough R.H. & Ogden R.G. (1993) Building design using cold formed steel sections:
Acoustic insulation. The Steel Construction Institute, Ascot, Berks.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Further reading
221
Grubb P.J. & Lawson R.M. (1997) Building design using cold formed steel sections:
Construction detailing and practice. The Steel Construction Institute, Ascot, Berks.
Grubb P.J., Gorgolewski M.T. & Lawson R.M. (2001) Building design using cold
formed steel sections: Light steel framing in residential construction. The Steel
Construction Institute, Ascot, Berks.
Lawson R.M. (1993) Building design using cold formed steel sections: Fire protection. The Steel Construction Institute, Ascot, Berks.
Lawson R.M., Grubb P.J., Prewer J. & Trebilcock P. (1999) Building design using
modular construction:An architect’s guide.The Steel Construction Institute,Ascot,
Berks.
Rogan A.L. & Lawson R.M. (1998) Value and benefit assessment of light steel
framing in housing. The Steel Construction Institute, Ascot, Berks.
Trebilcock P.J. (1994) Building design using cold formed steel sections: An architect’s
guide. The Steel Construction Institute, Ascot, Berks.
Section 5.5
Baker N. (1983) Atria and conservatories, Part 1. Architect’s Journal, 177, No. 20, 18
May, 67–9, and Part 2. Architect’s Journal, 177, No. 21, 25 May, 67–70.
Bednar M.J. (1986) The New Atrium. McGraw-Hill.
Chartered Institution of Building Services Engineers (CIBSE) (1999) CIBSE Guide
A Environmental Design, 6th Edn. CIBSE.
DeCicco P.R. (1983) Life Safety Considerations in Atrium Buildings. Fire Prevention 164, November, 27–33. Polytechnic Institute NY.
Dickson M.G.T. & Green M.G. (1986) Providing Intermediate Space. National Structural Steel Conference. BCSA.
Hawkes D. (1983) Atria and conservatories, Part 1. Architect’s Journal, 177, No. 19,
11 May, 67–70.
Kendrick C., Martin A.J. & Booth W. (1998) Refurbishment of air-conditioned
buildings for natural ventilation, Technical Note TN8/98, The Building Services
Research and Information Association.
Land District Surveyors Association (1989) Fire Safety in Atrium Buildings.
Law M. & O’Brien T. (1989) Fire Safety of Bare External Structural Steel. Steel
Construction Institute, Ascot, Berks.
Lloyds Chambers (1983) Framed in Steel, No. 11, Nov., Corus.
Martin A.J. (1995) Control of natural ventilation, Technical Note TN11/95, The
Building Services Research and Information Association.
Steel Designers' Manual - 6th Edition (2003)
Chapter 6
Applied metallurgy of steel
by MICHAEL BURDEKIN
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
6.1 Introduction
The versatility of steel for structural applications rests on the fact that it can be
readily supplied at a relatively cheap price in a wide range of different product
forms, and with a useful range of material properties. The key to understanding the
versatility of steel lies in its basic metallurgical behaviour. Steel is an efficient material for structural purposes because of its good strength-to-weight ratio. A diagram
of strength-to-weight ratio against cost per unit weight for various structural materials is shown in Fig. 6.1. Steel can be supplied with strength levels from about
250 N/mm2 up to about 2000 N/mm2 for common structural applications, although
the strength requirements may limit the product form. The material is normally
ductile with good fracture toughness for most practical applications. Product forms
range from thin sheet material, through optimized structural sections and plates, to
heavy forgings and castings of intricate shape. Although steel can be made to a wide
range of strengths it generally behaves as an elastic material with a high (and relatively constant) value of the elastic modulus up to the yield or proof strength. It also
usually has a high capacity for accepting plastic deformation beyond the yield
strength, which is valuable for drawing and forming of different products, as well as
for general ductility in structural applications.
Steel derives its mechanical properties from a combination of chemical composition, heat treatment and manufacturing processes. While the major constituent of
steel is always iron the addition of very small quantities of other elements can have
a marked effect upon the type and properties of steel. These elements also produce
a different response when the material is subjected to heat treatments involving
cooling at a prescribed rate from a particular peak temperature. The manufacturing process may involve combinations of heat treatment and mechanical working
which are of critical importance in understanding the subsequent performance of
steels and what can be done satisfactorily with the material after the basic manufacturing process.
Although steel is such an attractive material for many different applications, two
particular problems which must be given careful attention are those of corrosion
behaviour and fire resistance, which are dealt with in detail in Chapters 35 and 34
respectively. Corrosion performance can be significantly changed by choice of a steel
of appropriate chemical composition and heat treatment, as well as by corrosion
protection measures. Although normal structural steels retain their strength at tem222
Steel Designers' Manual - 6th Edition (2003)
Chemical composition
223
strength
weight
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
strength limited
/
//
/
/
/
/
/
/
/
//
/
x wood
//
ft steel
'oncrete
cost limited
ceromics (75 37)
titonium (433)
corbon fibre (160, 31)
x oluminium
x plostic
I
x copper
I
6
8
cost
10
weight
Fig. 6.1 Strength/weight and cost/weight ratios for different materials normalized to
steel (1,1)
peratures up to about 300°C, there is a progressive loss of strength above this temperature so that in an intense fire, bare steel may lose the major part of its structural strength. Although the hot strength and creep strength of steels at high
temperature can be improved by special chemical formulation, it is usually cheaper
to provide fire protection for normal structural steels by protective cladding.
6.2 Chemical composition
6.2.1 General
The key to understanding the effects of chemical composition and heat treatment
on the metallurgy and properties of steels is to recognize that the properties depend
upon the following factors:
(1)
(2)
(3)
(4)
(5)
microstructure
grain size
non-metallic inclusions
precipitates within grains or at grain boundaries
the presence of absorbed or dissolved gases.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
224
Applied metallurgy of steel
Steel is basically iron with the addition of small amounts of carbon up to a
maximum of 1.67% by weight, and other elements added to provide particular
mechanical properties. Above 1.67% carbon the material generally takes the form
of cast iron. As the carbon level is increased, the effect is to raise the strength level,
but reduce the ductility and make the material more sensitive to heat treatment.
The cheapest and simplest form is therefore a plain carbon steel commonly supplied for the steel reinforcement in reinforced concrete structures, for wire ropes,
for some general engineering applications in the form of bars or rods, and for some
sheet/strip applications. However, plain carbon steels at medium to high carbon
levels give rise to problems where subsequent fabrication/manufacturing takes
place, particularly where welding is involved, and more versatility can be obtained
by keeping carbon to a relatively low level and adding other elements in small
amounts. When combined with appropriate heat treatments, addition of these other
elements produces higher strength while retaining good ductility, fracture toughness, and weldability, or the development of improved hot strength, or improved
corrosion-resistance. The retention of good fracture toughness with increased
strength is particularly important for thick sections, and for service applications at
low temperatures where brittle fracture may be a problem. Hot strength is important for service applications at high temperatures such as pressure vessels and piping
in the power generation and chemical process plant industries. Corrosion-resistance
is important for any structures exposed to the environment, particularly for structures immersed in sea water. Weathering grades of steel are designed to develop a
tight adherent oxide layer which slows down and stifles continuing corrosion under
normal atmospheric exposure of alternate wet and dry conditions. Stainless steels
are designed to have a protective oxide surface layer which re-forms if any damage
takes place to the surface, and these steels are therefore designed not to corrode
under oxidizing conditions. Stainless steels find particular application in the chemical industry.
6.2.2 Added elements
The addition of small amounts of carbon to iron increases the strength and the sensitivity to heat treatment (or hardenability, see later). Other elements which also
affect strength and hardenability, although to a much lesser extent than carbon, are
manganese, chromium, molybdenum, nickel and copper. Their effect is principally
on the microstructure of the steel, enabling the required strength to be obtained for
given heat treatment/manufacturing conditions, while keeping the carbon level very
low. Refinement of the grain structure of steels leads to an increase in yield strength
and improved fracture toughness and ductility at the same time, and this is therefore an important route for obtaining enhanced properties in steels. Although heat
treatment and in particular cooling rate are key factors in obtaining grain refinement, the presence of one or more elements which promote grain refinement
by aiding the nucleation of new grains during cooling is also extremely beneficial.
Steel Designers' Manual - 6th Edition (2003)
Chemical composition
225
30
C
I',
0+
L)
c
+
z
C
4)
0
'110
4)
Q
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
C
0
10
20
30
40
chromium equivalent (Cr+Mo+ 1 .5Si+O.5Nb)
Fig. 6.2
Constitution (Schaeffler) diagram for stainless steels
Elements which promote grain refinement, and which may be added in small quantities up to about 0.050%, are niobium, vanadium and aluminium.
The major elements which may be added for hot strength and also for corrosionresistance are chromium, nickel and molybdenum. Chromium is particularly beneficial in promoting corrosion-resistance as it forms a chromium oxide surface layer
on the steel, which is the basis of stainless steel corrosion protection in oxidizing
environments. When chromium and nickel are added in substantial quantities with
chromium levels in the range 12% to 25%, and nickel content up to 20%, different
types of stainless steel can be made. As with the basic effect of carbon in the iron
matrix, certain other elements can have a similar effect to chromium or nickel but
on a lesser scale. From the point of view of the effects of chemical composition, the
type of stainless steel formed by different combinations of chromium and nickel can
be shown on the Schaeffler diagram in Fig. 6.2. The three basic alternative types of
stainless steel are ferritic, austenitic and martensitic stainless steels, which have different inherent lattice crystal structures and microstructures, and hence may show
significantly different performance characteristics.
6.2.3 Non-metallic inclusions
The presence of non-metallic inclusions has to be carefully controlled for particular applications. Such inclusions arise as a residue from the ore in the steelmaking
process, and special steps have to be taken to reduce them to the required level. The
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
226
Applied metallurgy of steel
commonest impurities are sulphur and phosphorus, high levels of which lead to
reduced resistance to ductile fracture and the possibility of cracking problems in
welded joints. For weldable steels the sulphur and phosphorus levels must be kept
less than 0.050%, and with modern steel-making practice should now preferably be
less than 0.010%. They are not always harmful however and in cases where welding
or fracture toughness are not important, deliberate additions of sulphur may be
made up to about 0.15% to promote free machining qualities of steel, and small
additions of phosphorus may be added to non-weldable weathering grade steels.
Other elements which may occur as impurities and may sometimes have serious
detrimental effects in steels are tin, antimony and arsenic, which in certain steels
may promote a problem known as temper embrittlement, in which the elements
migrate to grain boundaries if the steel is held in a temperature range between about
500°C and 600°C for any length of time. At normal temperature steels in this condition can have very poor fracture toughness, with failure occurring by inter-granular fracture. It is particularly important to ensure that this group of tramp elements
is eliminated from low alloy steels.
Steels with a high level of dissolved gases, particularly oxygen and nitrogen, can
behave in a brittle manner. The level of dissolved gases can be controlled by addition of small amounts of elements with a particular affinity for them so that the
element combines with the gas and either floats out in the liquid steel at high temperature or remains as a distribution of solid non-metallic inclusions. A steel with
no such additions to control oxygen level is known as a rimming steel, but for most
structural applications the elements silicon and/or aluminium are added as deoxidants. Aluminium also helps in controlling the free nitrogen level, which it is important to keep to low levels in cases where the phenomenon of strain ageing
embrittlement may be important.
6.3 Heat treatment
6.3.1 Effect on microstructure and grain size
During the manufacture of steel the required chemical composition is achieved
while it is in the liquid state at high temperature. As the steel cools, it solidifies at
the melting temperature at about 1350°C, but substantial changes in structure take
place during subsequent cooling and may also be affected by further heat treatments. If the steel is cooled slowly, it is able to take up the equilibrium type of lattice
crystal structure and microstructure appropriate to the temperature and chemical
composition.
These conditions can be summarized on a phase or equilibrium diagram for the
particular composition; the equilibrium diagram for the iron–iron carbide system is
shown in Fig. 6.3. Essentially this is a diagram of temperature against percentage of
carbon by weight in the iron matrix. At 6.67% carbon, an inter-metallic compound
called cementite is formed, which is an extremely hard and brittle material. At the
Steel Designers' Manual - 6th Edition (2003)
Heat treatment
227
1600
L+ ô
1400
liquid
L
1200
(-)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
a
1000
eutectic
E
V
800
600
400
0
1
2
3
4
5
6
7
weight % carbon
Fig. 6.3 Equilibrium phase diagram for iron – iron carbide system (f.c.c., face-centred cubic;
b.c.c., body-centred cubic)
left-hand end of the diagram, with very low carbon contents, the equilibrium structure at room temperature is ferrite. At carbon contents between these limits the
equilibrium structure is a mixture of ferrite and cementite in proportion depending
on the carbon level. On cooling from the melting temperature, at low carbon levels
a phase known as delta ferrite is formed first, which then transforms to a different
phase called austenite. At higher carbon levels, the melting temperature drops with
increasing carbon level and the initial transformation may be direct to austenite.
The austenite phase has a face-centred cubic lattice crystal structure, which is maintained down to the lines AE and BE on Fig. 6.3. As cooling proceeds slowly the
austenite then starts to transform to the mixture of ferrite and cementite which
results at room temperature. However, point E on the diagram represents a eutectoid at a composition of 0.83% carbon at which ferrite and cementite precipitate
alternately in thin laths to form a structure known as pearlite. At compositions less
than 0.83% carbon, the type of microstructure formed on slow cooling transformation from austenite is a mixture of ferrite and pearlite. Each type of phase present
at its appropriate temperature has its own grain size, and the ferrite/pearlite grains
tend to precipitate in a network within and based on the previous austenite grain
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
228
Applied metallurgy of steel
boundary structure. The lattice crystal structure of the ferrite material which forms
the basic matrix is essentially a body-centred cubic structure. Thus in cooling from
the liquid condition, complex changes in both lattice crystal structure and
microstructure take place dependent on the chemical composition. For the equilibrium diagram conditions to be observed, cooling must be sufficiently slow to allow
time for the transformations in crystal structure and for the diffusion/migration of
carbon to take place to form the appropriate microstructures.
If a steel is cooled from a high temperature and held at a lower constant temperature for sufficient time, different conditions may result; these are represented
on a diagram known as the isothermal transformation diagram. The form of the
diagram depends on the chemical analysis and in particular on the carbon or related
element content. In plain carbon steels the isothermal transformation diagram typically has the shape of two letters C each with a horizontal bottom line as shown in
Fig. 6.4. The left-hand/upper curve on the diagram of temperature against time represents the start of transformation, and the right-hand/lower curve represents the
800
•stort
700
—
— _feit+ pearlite
600
:::
bainite
300
200
100
0
1
10
106
time (s)
Fig. 6.4
Isothermal transformation diagram for 0.2% C, 0.9% Mn steel
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Heat treatment
229
completion of transformation with time. For steels with a carbon content below
the eutectoid composition of 0.83% carbon, holding at a temperature to produce
isothermal transformation through the top half of the letter C leads to the formation of a ferrite/pearlite microstructure. If the transformation temperature is
lowered to pass through the lower part of the C curves, but above the bottom horizontal lines, a new type of microstructure is obtained, which is called bainite, which
is somewhat harder and stronger than pearlite, but also tends to have poorer fracture toughness. If the transformation temperature is dropped further to lie below
the two horizontal lines, transformation takes place to a very hard and brittle substance called martensite. In this case the face-centred cubic lattice crystal structure
of the austenite is not able to transform to the body-centred cubic crystal structure
of the ferrite, and the crystal structure becomes locked into a distorted form known
as a body-centred tetragonal lattice. Bainite and martensite do not form on equilibrium cooling but result from quenching to give insufficient time for the equilibrium transformations to take place.
The position and shape of the C curves on the time axis depend on the chemical
composition of the steel. Higher carbon contents move the C curve to the right on
the time axis, making the formation of martensite possible at slower cooling rates.
Alloying elements change the shape of the C curves, and an example for a low-alloy
steel is shown in Fig. 6.5. Additional effects on microstructure, grain size and result-
800
700
600
'1)
0
500
400
\
300
200
\mortbQsite
100
0
1
10
10
iü
time (s)
Fig. 6.5 Continuous cooling transformation diagram for 0.4% C, 0.8% Mn, 1% Cr, 0.2% Mo
steel
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
230
Applied metallurgy of steel
ant properties can be obtained by combinations of mechanical work at appropriate
temperatures during manufacture of the basic steel.
In addition to the effect of cooling rate on microstructure, the grain size is significantly affected by time at high temperatures and subsequent cooling rate. Long
periods of time at higher temperatures within a particular phase lead to the merging
of the grain boundaries and growth of larger grains. For ferritic crystal structures,
grain growth starts at temperatures above about 600°C, and hence long periods
in the temperature range 600°C to 850°C with slow cooling will tend to promote
coarse grain size ferrite/pearlite microstructures. Faster cooling through the upper
part of the C curves will give a finer grain structure but still of ferrite/pearlite
microstructure.
The type of microstructure present in a steel can be shown and examined by the
preparation of carefully polished and etched samples viewed through a microscope.
Etching with particular types of reagent attacks different parts of the microstructure preferentially, and the etched parts are characteristic of the type of microstructure. Examples of some of the more common types of microstructure mentioned
above are shown in Fig. 6.6. The basic microstructure of the steel is usually shown
by examination in the microscope to magnifications of from 100 to about 500 times.
Where it is necessary to examine the effects of very fine precipitates or grain boundary effects, it may be necessary to go to higher magnifications. With the electron
microscope it is possible to reach magnifications of many thousands and, with specialized techniques, to reach the stage of seeing dislocations and imperfections in
the crystal lattice itself.
6.3.2 Heat treatment in practice
In practical steelmaking or fabrication procedures cooling occurs continuously from
high temperatures to lower temperatures. The response of the steel to this form of
cooling can be shown on the continuous cooling transformation diagram (CCT
diagram) of Fig. 6.7. This resembles the isothermal transformation diagram, but the
effect of cooling rate can be shown by lines of different slopes on the diagram. For
example, slow cooling, following line (a) on Fig. 6.7, passes through the top part of
the C curve and leads to the formation of a ferrite/pearlite mixture. Cooling at an
intermediate rate, following line (b), passes through pearlite/ferrite transformation
at higher temperatures, but changes to bainite transformation at lower temperatures
so that a mixture of pearlite and bainite results. Rapid cooling following line (c)
misses the C curves completely and passes through the two horizontal lines to show
transformation to martensite. Thus in practice for any given composition of steel
different microstructures and resultant properties can be produced by varying the
cooling rate.
The microstructure and properties of a steel can be changed by carefully chosen
heat treatments after the original manufacture of the basic product form. A major
group of heat treatments is effected by heating the steel to a temperature such that
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
Heat treatment
231
Fig. 6.6 Examples of common types of microstructure in steel (magnitication ¥ 500) (courtesy of Manchester Materials Science Centre, UMIST)
Steel Designers' Manual - 6th Edition (2003)
Applied metallurgy of steel
232
l\ \
Ii \
ii "
800
j \aus'tenite
start
700
• finish
0
600
0
500
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
a)
E
.3
\
400
\
\( a)
300
\\
200
100
ferrite
+pearlite
0
1
10
102
103
10
10
106
time (s)
Fig. 6.7
Continuous cooling transformation diagram for 02% C, 0.9% Mn steel
it transforms back to austenite, this temperature being normally in the range 850°C
to 950°C. It is important to ensure that the temperature is sufficient for full transformation to austenite, otherwise a very coarse-grained ferritic structure may result.
It is also important that the austenitizing temperature is not too high, and that the
time at this temperature is not too long, otherwise a coarse-grained austenite structure will form, making subsequent transformation to fine grains more difficult. A
heat treatment in which cooling is slow and essentially carried out in a furnace is
known as annealing. This tends to lead to a relatively coarse-grained final structure,
as predicted by the basic equilibrium phase diagram, and is used to put materials
into their softest condition. If the steel is allowed to cool freely in air from the
austenitizing temperature, the heat treatment is known as normalizing, which gives
a finer grain size and hence tends to higher yield strength and better toughness for
a given composition of steel. Normalizing may be combined with rolling of a particular product form over a relatively narrow band of temperatures, followed by
natural cooling in air, in which case it is known as controlled rolling. When the steel
product form is cooled more rapidly by immersing it directly into oil or water, the
heat treatment is known as quenching. Quenching into a water bath is generally
more severe than quenching into an oil bath.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Manufacture and effect on properties
233
A second stage heat treatment to temperatures below the austenitizing range is
frequently applied, known as tempering. This has the effect of giving more time for
the transformation processes which were previously curtailed to develop further,
and can permit changes in the precipitation of carbides, allowing them to merge
together and develop into larger or spheroidal forms. These thermally activated
events are highly dependent on temperature and time for particular compositions.
The net effect of tempering is to soften previously hardened structures and make
them tougher and more ductile.
Both plain carbon and low alloy steels can be supplied in the quenched and tempered condition for plates and engineering sections to particular specifications. The
term ‘hardenability’ is used to describe the ability of steel to form martensite to
greater depths from the surface, or greater section sizes. There are, therefore,
practical limits of section thickness or size at which particular properties can be
obtained.
In BS 970 (some sections of which have been replaced by BS EN standards as
part of a phased transition), a range of compositions of engineering steels is given,
together with the choice of heat treatments and limiting section sizes for which different properties can be supplied. The heat treatment condition is represented by a
letter in the range P to Z. The more commonly supplied conditions are in the range
P to T. It should be noted that the term ‘hardenability’ does not refer to the absolute
hardness level which can be achieved, but to the ability to develop uniform hardening throughout the cross section. Cooling rates vary at different positions in the
cross section as heat is conducted away in a quenching operation from the surface.
It is sometimes necessary to apply heat treatment to components or structures
after fabrication, particularly when they have been welded. The aim is mainly to
relieve residual stresses but heat treatment may also be required to produce controlled metallurgical changes in the regions where undesirable effects of welding
have occurred. Applications at high temperatures may also lead to metallurgical
changes taking place in service. It is vitally important that where any form of heat
treatment is applied the possible metallurgical effects on the particular type of steel
are taken into account.
Heat treatments are sometimes applied to produce controlled changes in shape
or correction of distortion and again temperatures and times involved in these heat
treatments must be carefully chosen and controlled for the particular type of steel
being used.
6.4 Manufacture and effect on properties
6.4.1 Steelmaking
Manufacture of steel takes place mainly in massive integrated steelworks. The first
stage starts with iron ore and coke, which are mixed and heated to produce a sinter.
This mixture then has limestone added to form the burden or raw material fed into
a blast furnace. Reactions which take place at high temperature in the blast furnace
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
234
Applied metallurgy of steel
lead to the formation of iron; the molten iron is tapped continuously from the
bottom of the blast furnace. The molten metal at this stage is approximately 90%
to 95% iron, the remainder being impurities, which have to he removed or reduced
to acceptable levels at the next stage, that of steelmaking. This material is fed
together with recovered scrap iron or steel into the steelmaking furnace, the
common types of which are known as either a basic oxygen furnace or an electric
arc furnace. In the basic oxygen furnace oxygen is blown on to the molten metal by
a water-cooled lance. In the electric arc furnace heat is produced by an arc between
electrodes over the metal surface and the molten metal itself conducting electricity.
Chemical reactions take place following additions of selected materials to the
molten metal, which lead to the reduction of the impurities and to the achievement
of the required controlled chemical composition of the steel. The impurities are
reduced by addition of elements which combine and float out to the surface of the
molten metal in the slag or dross waste material on the surface. Deoxidation or
killing of the steel takes place in the final stages before the furnace is tapped. Older
steel manufacturing practice was to tap the steel from the furnace into ladies and
then pour the molten steel into large moulds to produce ingots. These ingots would
normally be allowed to solidify and cool before reprocessing at a later stage by
rolling into the required product form. Modern steelmaking practice has now moved
much more to a process known as continuous casting, in which molten steel is poured
at a steady rate into a mould to form a continuous solid strand from which lengths
of semi-finished product are cut for subsequent processing. Semi-finished products
take the form of slabs, billets or blooms. Continuous casting has the advantage of
eliminating the reheating and first stage rolling required in the ingot production
route, and is generally more efficient, but ingot production is still required for some
product forms.
6.4.2 Casting and forging
If the final product form is a casting the liquid steel is poured direct into a mould
of the required geometry and shape. Steel castings provide a versatile way of achieving the required finished product, particularly where either many items of the same
type are required and/or complex geometries are involved. Special skills are
required in the design and manufacture of the moulds in order to ensure that good
quality castings are obtained with the required mechanical properties and freedom
from significant imperfections or defects. High-integrity castings for structural applications have been successfully supplied for critical components in bridges, such as
the major cable saddles for suspension bridges, cast node and tubular sections for
offshore structures, and the pump bowl casings for pressurized water reactor
systems. The size of component which can be made in cast form is limited to a
maximum of some 30–50 t however, and only a small proportion of total steel production is completed as castings for direct application.
Steel Designers' Manual - 6th Edition (2003)
Manufacture and effect on properties
235
Another specialist route to the finished steel product is by forging, in which a
bloom is heated to the austenitizing temperature range and formed by repeated
mechanical pressing in different directions to achieve the required shape. The combination of temperature and mechanical work enables high-quality products with
good mechanical properties to be obtained. An example of high-integrity forgings
is the production of steel rings to form the shell/barrel of the reactor pressure vessel
in a pressurized water reactor system. Again the proportion of steel production as
forgings is a relatively small and specialized part of overall steel production.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
6.4.3 Rolling
By far the largest amount of finished steel production is achieved by rolling. The
semi-finished products cast from the steelmaking furnace are reheated to the
austenitizing range and passed through a series of mills with rolls of the required
profile to force the hot steel into the finished shape. An example of the distribution
of steel products supplied to the UK construction industry is shown in the pie chart
of Fig. 6.8. It can be seen that a major part is strip or sheet material which is produced by continuous rolling from slabs down to sheet of the required width and
thickness which is first collected as a coil at the end of the rolling process, and subsequently cut into required lengths. The sheet material can be supplied either in
bare steel form or with different types of coating. For example, it is possible to obtain
steel sheet with a continuous galvanizing (zinc) coating for corrosion protection and
it is also possible to obtain it with integral plastic coatings of different colours and
patterns for decorative finish as well as corrosion protection.
Fig. 6.8 Supply of steel products to the UK construction industry 2001 (UK Steel
Association)
Steel Designers' Manual - 6th Edition (2003)
236
Applied metallurgy of steel
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
For the structural industry steel slabs can be rolled into plates of the required
thickness, or into structural sections such as universal columns, universal beams,
angle sections, rail sections, etc. Round blooms or ingots can be processed by a seamless tube rolling mill into seamless tubes of different diameters and thicknesses or
solid bar subsequently drawn out into wire. Tubes can be used either for carrying
fluids in small-diameter pipelines or as structural hollow sections of circular or
rectangular shape. The shape of engineering structural sections is determined by
the required properties of the cross section such as cross-sectional area and
moments of inertia about different axes to give an effective distribution of the
weight of the material for structural purposes. Rolled structural sections are supplied in a standard range of shapes detailed in BS 4: Part 1, and a selection of typical
shapes and section properties is given in the Appendix Geometrical properties of
plane sections.
6.4.4 Defects
In any bulk manufacturing process, such as the manufacture of steel, it is inevitable
that a small proportion of the production will have imperfections which may or may
not be harmful from the point of view of intended service performance of the
product. In general the appropriate applications standards have clauses which limit
any such imperfections to acceptable and harmless levels. In castings a particular
family of imperfections can occur which are dependent on the material and the
geometry being manufactured. The most serious types of imperfection are cracks
caused by shrinkage stresses during cooling, particularly at sharp changes in cross
section. A network of fine shrinkage cracks or tears can sometimes develop, again
particularly at changes in cross section where the metal is subjected to a range of
different cooling rates. The second type of imperfection in castings is solid inclusions, particularly in the form of sand where this medium has been used to form the
moulds. Porosity, or gaseous inclusions, is not uncommon in castings to some degree
and again tends to occur at changes in cross section. There is usually appreciable
tolerance for minor imperfections such as sand inclusions or porosity provided these
do not occur to extreme levels.
In rolled or drawn products, the most common types of defect are either cold laps
or rolled-in surface imperfections. A lap is an imperfection which forms when the
material has been rolled back on to itself but has not fully fused at the interface.
Surface imperfections may occur from the same cause where a tongue of material
is rolled down but does not fuse fully to the underlying material. Both of these faults
are normally superficial and in any serious cases ought to be eliminated by final
inspection at the steel mills. A third form of imperfection which can occur in plates,
particularly when produced from the ingot route, is a lamination: the failure of the
material to fuse together, usually at the mid-thickness of the plate. Laminations tend
to arise from the rolling-out of pipes, or separation on the centreline of an ingot
at either top or bottom which formed at the time of casting the ingot. Normal
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Engineering properties and mechanical tests
237
practice is that sufficient of the top and bottom of an ingot is cut off before subsequent processing to prevent laminations being rolled into subsequent products, but
nevertheless they do occur from time to time. Fortunately the development of cracks
in rolled products is relatively rare although it may occasionally occur in drawn
products or as a result of quenching treatments in heat-treated products.
Since much of the manufacture of steels involves processing at, and subsequent
cooling from, high temperatures it will be appreciated that high thermal stresses can
develop during differential cooling and these can lead to residual stresses in the
finished product. In many cases these residual stresses are of no significance to the
subsequent performance of the product but there are situations where their effect
must be taken into account. The two in which residual stresses from the steel manufacture are most likely to be of importance are where close tolerance machining is
required, or where compression loading is being applied to slender structural sections. For the machining case it may be necessary to apply a stress relief treatment,
or alternatively to carry out the machining in a series of very fine cuts. The effect
of the inherent manufacturing residual stresses on structural sections is taken
account of in the design codes such as BS 5950 for steel buildings, by giving a varying
factor on the limiting permissible stresses depending on the product shape. These
factors have been determined by a series of research programmes on the buckling
behaviour of different shaped sections coupled to measurements of the inherent
residual stresses present.
6.5 Engineering properties and mechanical tests
As part of the normal quality control procedures of the steel manufacturer, and as
laid down in the different specifications for manufacture of steel products, tests are
carried out on samples representing each batch of steel and the results recorded on
a test certificate. At the stage when the chemical analysis of the steel is being
adjusted in the steelmaking furnace, samples are taken from the liquid steel melt at
different stages to check the analysis results. Samples are also taken from the melt
just before the furnace is tapped, and the analysis of these test results is taken to
represent the chemical composition of the complete cast. The results of this analysis are given on test certificates for all products which are subsequently made from
the same initial cast. The test certificate will normally give analysis results for C, Mn,
Si, S and P for all steels, and where the specification requires particular elements to
be present in a specific range the results for these elements will also be given. Even
when additional elements are not specified, the steel manufacturer will often
provide analysis results for residual elements which may have been derived from
scrap used or which could affect subsequent fabrication of performance
during fabrication particularly welding. Thus steel supplied to the Specification for
weldable structural steels, BS 4360, will often have test certificates giving Cr, Ni,
Cu, V, Mb and Al, as well as the main basic five elements.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
238
Applied metallurgy of steel
In some specifications, the requirement is given for additional chemical analysis
testing on each item of the final product form, and this is presented on the test certificates as product analysis in addition to the cast analysis. This does however incur
additional costs. The Specification for weldable structural steels gives the opportunity for requiring the steelmaker to supply information on the carbon equivalent to
assist the fabricator on deciding about precautions during welding (see later). In
low-alloy and stainless steels, the test certificates will of course give the percentage
of the alloying elements such as chromium, nickel, etc.
The test certificates should also give the results of mechanical tests on samples
selected to represent each product range in accordance with the appropriate specification. The mechanical test results provided will normally include tensile tests
giving the yield strength, ultimate strength and elongation to failure. In structural
steels, where the fracture toughness is important, specifications include requirements for Charpy V-notch impact tests to BS 131: Part 2. The Charpy test is a standard notched bar impact test of 10 mm square cross section with a 2 mm deep
V-notch in one face. A series of specimens is tested under impact loading either at
one specification temperature or over a range of temperatures, and the energy
required to break the sample is recorded. In the Euronorms, these notch ductility
requirements are specified by letter grades JR, JO, J2 and K2. Essentially these
requirements are that the steel should show a minimum of 27 J energy absorption
at a specified testing temperature corresponding to the letter grade.
Charpy test requirements are also included in some of the general engineering
steel specifications (BS 970) for pressure vessel steels and other important structural applications, particularly where welded structures are used.
The specifications normally require the steelmaker to extract specimens with their
length parallel to the main rolling direction. In fact it is unlikely that the steel will
be wholly isotropic, and significant differences in material properties may occur
under different testing directions, which would not be evident from the normal test
certificates unless special tests were carried out. It is possible in some specifications
to have material tests carried out both transverse to the main rolling direction, and
in the through-thickness direction of rolled products. Testing in the throughthickness direction is particularly important where the material may in fact be
loaded in this direction in service by welded attachments. Since such tests are additional to the normal routine practice of the steel manufacturer, and cost extra both
for the tests themselves and for the disruption to main production, it is not unexpected that steels required to be tested to demonstrate properties in other directions are more expensive than the basic quality of steel tested in one standard
direction only. The quality-control system at the steel manufacturers normally puts
markings in the form of stamped numbers or letters on each length or batch of products so that it can be traced back to its particular cast and manufacturing route. In
critical structural applications it is important that this numbering system is transferred on through fabrication to the finished structure so that each piece can be
identified and confirmed as being of the correct grade and quality. The test certificate for each batch of steel is therefore a most important document to the steel
manufacturer, to the fabricator, and to the subsequent purchaser of the finished
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fabrication effects and service performance
239
component or structure. In addition to the chemical composition and mechanical
properties, the test certificate should also record details of the steelmaking route
and any heat treatments applied to the material by the steel manufacturer.
It is not uncommon for some semi-finished products to be sold by the steel manufacturer to other product finishers, or to stockholders. Unless these parties retain
careful records of the supply of the material it may be difficult to trace specific
details of the properties of steel bought from them subsequently, although some
stockholders do maintain such records.
Where products are manufactured from semi-finished steel and subsequently
given heat treatment for sale to the end user, the intermediate manufacturer should
produce his own test certificates detailing both the chemical analysis of the steel
and the mechanical properties of the finished product. For example, bolts used for
structural connections are manufactured from bar material and are normally
stamped with markings indicating the grade and type of bolt. Samples of bolts are
taken from manufactured batches after heat treatment and subjected to mechanical tests to give reassurance that the correct strength of steel and heat treatment
have been used.
6.6 Fabrication effects and service performance
Basic steel products supplied from the steel manufacturer are rarely used directly
without some subsequent fabrication. The various processes involved in fabrication
may influence the suitability for service of the steel, and over the years established
procedures of good practice have been developed which are acceptable for particular industries and applications.
6.6.1 Cutting, drilling, forming and drawing
Basic requirements in the fabrication of any steel component are likely to be cutting
and drilling. In thin sections, such as sheet material, steel can be cut satisfactorily
by guillotine shearing, and although this may form a hardened edge it is usually of
little or no consequence. Thicker material in structural sections up to about 15 mm
thickness can also be cut by heavy-duty shears, useful for small part pieces such as
gussets, brackets, etc. Heavier section thicknesses will usually have to be cut by cold
saw or abrasive wheel or by flame cutting. Cold saw and abrasive wheel cutting
produce virtually no detrimental effects and give good clean cuts to accurate dimensional tolerances. Flame cutting is carried out using an oxyacetylene torch to burn
the steel away in a narrow slit, and this is widely used for cutting of thicker sections
in machine-controlled cutting equipment. The intense heating in flame cutting does
subject the edge of the metal to rapid heating and cooling cycles and so produces
the possibility of a hardened edge in some steels. This can be controlled by either
preheating just ahead of the cutting torch or using slower cutting speeds, or
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
240
Applied metallurgy of steel
alternatively, if necessary, any hardened edge can be removed by subsequent
machining. In recent years laser cutting has become a valuable additional cutting
method for thin material, in that intricate shapes and patterns can be cut out rapidly
by steering a laser beam around the required shape.
Drilling of holes presents little problem and there are now available numerical/computer controlled systems which will drill multiple groups of holes to the
required size and spacing. For thinner material hole punching is commonly used and
although this, like shearing, can produce a hardened edge, provided the punch is
sharp no serious detrimental effects occur in thinner material.
It is sometimes necessary to bend, form or draw steel into different shapes. Reinforcing steel for reinforced concrete structures commonly has to be bent into the
form of hooks and stirrups. The curved sections of tubular members of offshore
structures or cylindrical parts of pressure vessels are often rolled from flat plate to
the required curvature. In these cases yielding and plastic strain take place as the
material is deformed beyond its elastic limit. This straining moves the material condition along its basic stress–strain curve, and it is therefore important to limit the
amount of plastic strain used up in the fabrication process so that that availability
for subsequent service is not diminished to an unacceptable extent. The important
variable in limiting the amount of plastic strain which occurs during cold forming
is usually the ratio of the radius of any bend to the thickness or diameter of the
material. Provided this ratio is kept high the amount of strain will be limited. Where
the amount of cold work which has been introduced during fabrication is excessive,
it may be necessary to carry out a reheat treatment in order to restore the condition of the material to give its required properties.
In the manufacture of wire, the steel is drawn through a series of dies gradually
reducing its diameter and increasing the length from the initial rod sample. This cold
drawing is equivalent to plastic straining and has the effect of both increasing the
strength of the material and reducing its remaining ductility as the material moves
along its stress/strain curve. In certain types of wire manufacture, intermediate heat
treatments are necessary in order to remove damaging effects of cold work and
enhance and improve the final mechanical properties.
6.6.2 Welding
One of the most important fabrication processes for use with steel is welding. There
are many different types of welding and this subject is itself a fascinating multi-disciplinary world involving combined studies in physics, chemistry, electronics, metallurgy, and mechanical, electrical and structural engineering. Most welding processes
involve fusion of the material being joined, by raising the temperature to the melting
point of the material, either with or without the addition of separate filler metal.
Although there is a huge variety of different welding processes, probably the most
common and most important ones for general applications are the group of arc
welding processes and the group of resistance welding processes. Among the newer
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fabrication effects and service performance
241
processes are the high energy density beam processes such as electron beam and
laser welding.
Arc welding processes involve the supply of an intense heat source from an electric arc which melts the parent material locally, and may provide additional filler
metal by the melting of a consumable electrode. These processes are extensively
used in the construction industry, and for any welding of material thicknesses above
the range of sheets. The resistance group of welding processes involve the generation of heat at the interface between two pieces of material by the passage of very
heavy current directly between opposing electrodes on each face. The resistance
processes do not involve additional filler metal, and can be used to produce local
joints as spot welds, or a series of such welds to form a continuous seam. This group
of processes is particularly suitable for sheet material and is widely used in the automotive and domestic equipment markets.
It will be appreciated that fusion welding processes involve rapid heating and
cooling locally at the position where a joint is to be made. The temperature gradients associated with welding are intense, and high thermal stresses and subsequent
residual stresses on cooling are produced. The residual stresses associated with
welding are generally much more severe than those which result during the basic
steel manufacturing process itself as the temperature gradients are more localized
and intense. Examples of the residual stress distribution resulting from the manufacture of a butt weld between two plates and a T-butt weld with one member
welded on to the surface of a second are shown in Fig. 6.9. As will be seen from
other chapters, residual stresses can be important in the performance of steel structures because of their possible effects on brittle fracture, fatigue and distortion. If
a steel material has low fracture toughness, and is operating below its transition temperature, residual stresses may be very important in contributing to failure by brittle
fracture at low applied stresses. If on the other hand the material is tough and yields
extensively before failure, residual stresses will be of little importance in the overall
structural strength. These effects are summarized in Fig. 6.10.
In fatigue-loaded structures residual stresses from welding are important in altering the mean stress and stress ratio. Although these are secondary factors compared
to the stress range in fatigue, the residual stress effect is sufficiently important that
it is now commonly assumed that the actual stress range experienced at a weld operates with an upper limit of the yield strength due to locked-in residual stresses
at this level. Thus although laboratory experiments demonstrate different fatigue
performance for the same stress range at different applied mean stress in plain
unwelded material, the trend in welded joints is for the applied stress ratio effect to
be overridden by locked-in residual stresses.
The effect of welding residual stresses on distortion can be significant, both at the
time of fabrication and in any subsequent machining which may be required. The
forces associated with shrinkage of welds are enormous, and will produce overall
shrinkage of components and bending/buckling deformations out of a flat plane.
These effects have to be allowed for either by pre-setting in the opposite direction
to compensate for any out-of-plane deformations or by making allowances with
components initially over-length to allow for shrinkage.
Steel Designers' Manual - 6th Edition (2003)
242
Applied metallurgy of steel
UR
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
((1 ( ( (( ((( ( (( ((( (((
LA
AA
(a)
B
BB
(b)
Fig. 6.9 Typical weld residual stress distributions: (a) butt weld, (b) T-butt weld (sR
residual stress; sy yield stress)
Steel Designers' Manual - 6th Edition (2003)
Fabrication effects and service performance
243
(b)
toughness
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(a)
temperature
load
load
/
displacement
(a) low toughness
Fig. 6.10
displacement
(b) high toughness
Effects of toughness and residual stresses on strength in tension
Just as the basic steel manufacturing process can lead to the presence of imperfections, welding also can lead to imperfections which may be significant. The types
of imperfection can be grouped into three main areas: planar discontinuities, nonplanar (volumetric) discontinuities, and profile imperfections. By far the most
serious of these are planar discontinuities, as these are sharp and can be of a
significant size. There are four main types of weld cracking which can occur in
steels as planar discontinuities. These are solidification (hot) cracking, hydrogeninduced (cold) cracking, lamellar tearing, and reheat cracking. Examples of these are
shown in Fig. 6.11. Hot cracking occurs during the solidification of a weld due to the
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
244
Applied metallurgy of steel
I.
:1
Fig. 6.11 Examples of different types of cracking which may occur in welded steel joints
(courtesy of The Welding Institute)
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Summary
245
rejection of excessive impurities to the centreline. Impurities responsible are usually
sulphur and phosphorus, and the problem is controlled by keeping them to a low
level and avoiding deep narrow weld beads. Cold cracking is due to the combination of a susceptible hardened microstructure and the effects of hydrogen in the
steel lattice. The problem is avoided by control of the steel chemistry, arc energy
heat input, preheat level, quenching effect of the thickness of joints being welded,
and by careful attention to electrode coatings to keep hydrogen potential to very
low levels. Guidance on avoiding this type of cracking in the heat affected zones of
weldable structural steels is given in BS 5135.
Lamellar tearing is principally due to the presence of excessive non-metallic inclusions in rolled steel products resulting in the splitting open of these inclusions under
the shrinkage forces of welds made on the surface. The non-metallic inclusions
usually responsible are either sulphides or silicates; manganese sulphides are probably the most common. The problem is avoided by keeping the impurity content
low, particularly the sulphur level to below about 0.010%, and by specifying tensile
tests in the through-thickness direction to show a minimum ductility by reduction
of area dependent on the amount of weld shrinkage anticipated (i.e. size of welded
attachment, values of R of A of 10% to 20% are usually adequate). Reheat cracking is a form of cracking which can develop during stress relief heat treatment or
during high temperature service in particular types of steel (usually molybdenum
or vanadium bearing) where secondary precipitation of carbides develops before
relaxation of residual stresses has taken place.
Other forms of planar defect in welds are the operator or procedure defects of
lack of penetration and lock of fusion. The volumetric/non-planar imperfections
divide into the groups of solid inclusions and gaseous inclusions. The solid inclusions are usually slag from the electrode/flux coating and the gaseous inclusions
result from porosity trapped during the solidification of the weld. In general the
non-planar defects are much less critical than planar defects of the same size and
are usually limited in their effect because their size is inherently limited by their
nature.
6.7 Summary
6.7.1 Criteria influencing choice of steel
The basic requirement in the choice of a particular steel is that it must be fit for
the product application and design conditions required. It must be available in the
product form and shape required and it should be at the minimum cost for
the required application. Clearly before the generic type of material is chosen
as steel it must be shown to be advantageous to use steel over other contending
materials, and therefore the strength-to-weight ratio and cost ratios must be
satisfactory.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
246
Applied metallurgy of steel
The steel must have the required strength, ductility and long-term service life in
the required environmental service conditions. For structural applications the steel
must also have adequate fracture toughness, this requirement being implemented
by standard Charpy test quality control levels.
Where the steel is to be fabricated into components or structures, its ability to
retain its required properties in the fabricated condition must be clearly established.
One of the most important factors in this respect for a number of industries is the
weldability of steel, and in this respect the chemical composition of the steel must
be controlled within tight limits, and the welding processes and procedures adopted
must be compatible with the material chosen.
The corrosion-resistance and potential fire-resistance/high temperature performance of the steel may be important factors in some applications. A clear decision
has to be taken at the design stage as to whether resistance to these effects is to be
achieved by external or additional protection measures, or inherently by the chemical composition of the steel itself. Stainless steels with high quantities of chromium
and nickel are significantly more expensive than ferritic carbon or carbon
manganese steels. Particular application standards generally specify the range of
material types which are considered suitable for their particular application.
Increased strength of steels can be obtained by various routes, including increased
alloying content, heat treatment, or cold working. In general as the strength
increases so does the cost and there may be little advantage in using high-strength
steels in situations where either fatigue or buckling are likely to be ruling modes of
failure. It should not be overlooked that although there is some increase in cost of
the basic raw material with increasing strength, there is likely to be a significant
increase in fabrication costs, with additional precautions necessary for the more
sophisticated types of higher-strength material.
Certain product forms are available only in certain grades of steel. It may not be
possible to achieve high strength in some product shapes and retain dimensional
requirements through the stage of heat treatment because of distortion problems.
Wherever possible, guidance should be sought on the basis of similar previous
experience or prototype trials to ensure that the particular material chosen will be
suitable for its required application.
6.7.2 Steel specifications and choice of grade
Structural steelwork, comprising rolled products of plate, sections and hollow
sections, is normally of a weldable carbon or carbon-manganese structural steel
to the new European based standard, BS EN 10025. Two strength grades are most
commonly used, grade S275 and grade S355, having yield strengths typically of
275 N/mm2 and 355 N/mm2 respectively. To help designers adjust from the old
BS 4360 deagnations, the Appendix Properties of steel contains a number of
tables which compare the old standard with the new European standards for
structural steels.
Further standards are cited in the Appendix British Standards for steelwork.
Steel Designers' Manual - 6th Edition (2003)
Further reading
247
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Further reading for Chapter 6
Baddoo N.R. & Burgon B.A. (2001) Structural Design in Stainless Steel. The Steel
Construction Institute, Ascot, Berks.
Dieter G.E. (1988) Mechanical Metallurgy, 3rd edn (SI metric edition). McGrawHill.
Gaskell D. (1981) Introduction to Metallurgical Thermodynamics, 2nd edn.
McGraw-Hill, New York.
Honeycombe R.W.K. (1995) Steels: Microstructure and Properties, 2nd edn. Edward
Arnold, London.
Lancaster J.F. (1987) Metallurgy of Welding, 4th edn. Allen and Unwin.
Porter D.A. & Easterling K.F. (2001) Phase Transformations in Metals and Alloys,
2nd edn. Nelson Thomas, Cheltenhum.
Smallman R.E. (1999) Modern Physical Metallurgy, 6th edn. ButterworthHeinemann, Oxford.
Szekely J. & Themelis N.J. (1971) Rate Phenomena in Process Metallurgy. Wiley.
Steel Designers' Manual - 6th Edition (2003)
Chapter 7
Fracture and fatigue
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
by JOHN YATES
Structural steelwork is susceptible to several failure processes. The principal
amongst these are wet corrosion, plastic collapse, fatigue cracking and rapid fracture, often termed brittle fracture. In this chapter, failure by rapid fracture and
fatigue cracking will be discussed.
7.1 Fracture
7.1.1 Introduction
The term brittle fracture is used to describe the fast, unstable fractures that occur
with very little energy absorption. In contrast, ductile fracture is a relatively slow
process that absorbs a considerable amount of energy, usually through plastic deformation. Some metals, such as copper and aluminium, have a crystalline structure
that enables them to resist fast fracture under all loading conditions and at all temperatures. This is not the case for many ferrous alloys, particularly structural steels,
which can exhibit brittle behaviour at low temperatures and ductile behaviour at
higher temperatures. The consequence of a brittle fracture in a structure may be an
unexpected, catastrophic failure. An understanding of the fundamentals of this
subject is therefore important for all structural engineers.
The introduction and development of fracture mechanics technology allows the
engineer to examine the susceptibility of steel structures, especially their welded
joints, to failure assuming that a defect of a given size is present and knowing the
operating conditions.
7.1.2 Ductile and brittle behaviour
Ductile fracture is normally preceded by extensive plastic deformation. Ductile
fracture is slow, and generally results from the formation and coalescence of voids.
These voids are often formed at inclusions due to the large tensile stresses set up
at the inclusion/metal interface, as seen in Fig. 7.1(a). Ductile fracture usually goes
through the grains but, if the density of inclusions or of pre-existing holes is higher
248
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
Fracture
transqranular
necking
Fig. 7.1
249
intergranular
(a)
shearing
(b)
Plastic deformation (a) by voids growth and coalescence, (b) by necking or
shearing
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
250
Fracture and fatigue
cleavage
Fig. 7.2
intergranular
Transgranular and intergranular brittle fracture
on grain boundaries than it is within the grains, then the fracture path may follow
the boundaries, giving a fibrous or ductile intergranular fracture. In cases where
inclusions are absent, it has been found that voids are formed in severely deformed
regions through localized slip bands and macroscopic instabilities, resulting in either
necking or the formation of zones of concentrated shear, as depicted in Fig. 7.1(b).
The fracture path of a ductile crack is often irregular, and the presence of a large
number of small voids gives the fracture surface a dull fibrous appearance.
The capacity of most metals of engineering interest for plastic deformation and
work hardening is extremely valuable as a safeguard against design oversight, accidental overloads or failure by cracking due to fatigue, corrosion or creep.
Brittle fracture is often thought to refer to rapid propagation of cracks without
any plastic deformation at a stress level below the yield stress of the material. In
practice, however, most brittle fractures show some, very limited, plastic deformation ahead of the crack tip. Brittle fracture may be transgranular (cleavage) or intergranular, as depicted in Fig. 7.2.
It is also worth mentioning that ferritic steels, which often show ductile behaviour, can, under certain circumstances, behave in a brittle fashion leading to fast
unstable crack growth. This has been clearly demonstrated over the years by some
unfortunate accidents involving ships, bridges, offshore structures, gas pipelines,
pressure vessels and other major constructions. The Liberty ships and the King
Street bridge in Melbourne, Australia, the Sea Gem drilling rig for North Sea gas
and the collapse of the Alexander Kielland oil rig are a few examples of the casualties of brittle fracture.
Steel Designers' Manual - 6th Edition (2003)
Fracture
251
200
., 150
U
'a
0.
E 100
0.
Ca
C)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
50
0.8% carbon
-100
-80
-60
-40
-20
0
20
40
60
80
100
120
140
160
Test temperature, °C
Fig. 7.3
The effect of temperature and carbon content on the impact energy of ferritic steels
An important feature of ferritic steels is the transition temperature between
ductile and brittle fracture. Understanding the factors which influence the transition temperature allows designers to be able to select a material which will be
ductile at the required operating temperatures for a given structure. The traditional
procedure for assessing the ductile to brittle transition in steels is by impact testing
small notched beams.1 The energy absorbed during the fracture process is a measure
of the toughness of the material and varies from a low value at low temperatures
to a high value as the temperature is raised. The characteristic shape of the impact
energy–temperature graph has led to the terminology of the upper and lower shelf.
The low temperature, brittle behaviour is often referred to as the lower shelf and
the high temperature, ductile behaviour the upper shelf.
The Charpy V-notch test1 is the most popular impact testing technique and is
described later in this chapter. The transition in impact toughness values obtained
from Charpy tests on carbon steels at different temperatures is shown in Fig. 7.3.
BS EN 100252 gives temperatures for minimum toughness values to be obtained for
a range of structural steel grades. Impact transition curves are a simple way of defining the effect that variables such as heat treatment, alloying elements and effects of
welding have on the fracture behaviour of a steel. Charpy values are useful for
quality control but more sophisticated tests3,4 are required if the full performance
of a material is to be exploited.
An understanding of the fracture behaviour of steel is particularly important
when considering welded structures. Welding can considerably reduce the toughness of plate in regions close to the fusion line and introduce defects in the weld
Steel Designers' Manual - 6th Edition (2003)
252
Fracture and fatigue
area. This, coupled with residual tensile stresses from the heating and cooling during
welding, can lead to cracking and eventual failure of a joint.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
7.2 Linear elastic fracture mechanics
The presence of a crack in a structure requires an assessment of whether the crack
is likely to grow by a fracture or fatigue mechanism.
The science of dealing with relatively large cracks in essentially elastic bodies is
linear elastic fracture mechanics. The assumptions upon which it is based are that
the crack is embedded in an isotropic, homogeneous, elastic continuum. In engineering practice, this means that a crack must be much larger than any microstructural feature, such as grain size, but it must be small in relation to the dimensions
of the structure, and the stresses present in the structure must be less than about
1/3 of the yield stress.
A crack in a solid can be opened in three different modes, as shown in Fig. 7.4.
The commonest is when normal stresses open the crack. This is termed mode I or
the opening mode. The essential feature of linear elastic fracture mechanics is that
all cracks adopt the same parabolic profile when loaded in mode I and that the
tensile stresses ahead of the crack tip decay as a function of 1/√(distance) from the
crack tip.
The absolute values of the opening displacements and the crack tip stresses
depend on the load applied and the length of the crack. The scaling factor for the
stress field and crack displacements is called the stress intensity factor, KI, and is
related to the crack length, a, and the remote stress in the body, s, by
K I = Ys pa
(7.1)
where Y is a geometric correction term to account for the proximity of the boundaries of the structure and the form of loading applied. The subscript I denotes the
mode I, or opening mode, of loading.
II
Fig. 7.4
Modes of crack opening
In
Steel Designers' Manual - 6th Edition (2003)
Linear elastic fracture mechanics
F
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
2a
2W
Fig. 7.5
253
F
a
W
Convention for describing the length of embedded and edge cracks
It is important to note that the convention is that an embedded crack that has
two tips has a length of 2a, and an edge crack which only has one tip has a length
of a: see Fig. 7.5. Values of the correction term Y have been compiled for many
geometries and load cases and are published in references 5 and 6. It is also
important to take care over the units used in calculating stress intensity factors.
Values may be quoted in MPa m, MN.m-1.5 or N.mm-1.5. The conversion between
them is
1 MPa m = 1 MN.m-1.5 = 31.62 N.mm-1.5
The usefulness of the stress intensity factor lies in the concept of similitude. That
is, if two cracks, one in a small laboratory specimen and one in a large structure,
have the same value of KI then they have identical opening displacements and identical crack tip stress fields. If the laboratory specimen fails in a brittle, catastrophic
manner at a critical value of KI then the structure will also fail when the stress intensity factor reaches that value. This critical value is a material property, termed the
fracture toughness, and is represented by KIc.
The practical applications of linear elastic fracture mechanics are in assessing the
likelihood that a particular combination of loading and crack size will cause a
sudden fracture. Given that a structure is at risk of failing if
Ys pa ≥ K Ic
(7.2)
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
254
Fracture and fatigue
then knowing two of the three parameters maximum stress, crack size and fracture
toughness will allow the limiting value of the third parameter to be determined.
For example, if the maximum allowable stress is 160 MPa and the maximum
allowable crack size in an edge cracked plate, where Y = 1.12, is 100 mm then
the minimum fracture toughness of the material must be 100 MPa m to avoid
fracture. (Remember that it is important to keep track of the units and 1 MPa m
= 31.62 N.mm-1.5.)
It has been found by experiment that the fracture toughness value measured in
a laboratory is influenced by the dimensions of the specimen. In particular, the
size and thickness of the specimen and the length of the remaining uncracked
section, called the ligament, control the level of constraint at the crack tip. Constraint is the development of a triaxial stress state which restricts the deformation
of the material. Thick specimens, which generate high constraint or plane strain
conditions, give lower values of the fracture toughness than those found from thin
specimens under low constraint or plane stress conditions. The least dimensions
to give the maximum constraint, and hence the minimum value of toughness, are
when:
Ê K Ic ˆ
thickness, width and ligament ≥ 2.5Á
˜
Ë sy ¯
2
(7.3)
This ensures that the localised plasticity at the crack tip is less than 2% of any of
the dimensions of the body and therefore does not disturb the elastic crack tip stress
field.
The size requirement of Equation (7.3) gives rise to practical problems in testing.
For example, steel with a room temperature yield strength of 275 N.mm-2 and a
minimum fracture toughness of 70 MPa m needs to be tested using a specimen at
least 160 mm thick to ensure plane strain conditions. In reality, much structural steelwork is made from sections substantially thinner than this and its fracture toughness will be significantly higher. It is good practice in many engineering disciplines
to measure fracture toughness values using specimens of similar thickness to the
proposed application.
Tables 3 to 7 in BS 5950: 20007 define, as a function of temperature, the maximum
thickness of steel that should be used to avoid brittle fracture. This is the same as
BS 5400-3: 20008 for steel bridge design. Both BS 5400 and BS 5950 are consistent
with, but generally more conservative than, the minimum thickness to ensure plane
strain conditions, as estimated from Equation (7.3). This means that the toughness
of the steel in use should be better than the minimum value achievable in thick
sections.
Linear elastic fracture mechanics is of much greater use on the lower shelf of
the ductile-brittle behaviour of steels than on the upper shelf. At low temperatures,
the yield strength is higher, the fracture toughness much lower and plane strain
conditions can be achieved in relatively thin sections. On the upper shelf, linear
elastic fracture mechanics tends to be inapplicable, and other techniques need to be
used.
Steel Designers' Manual - 6th Edition (2003)
Elastic–plastic fracture mechanics
255
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
7.3 Elastic–plastic fracture mechanics
The need to consider fracture resistance of materials outside the limits of validity
of plane strain linear elastic fracture mechanics (LEFM) is important for most engineering designs. To obtain valid KIc results for relatively tough materials it would
be necessary to use a test piece of dimensions so large that they would not be representative of the sections actually in use.
Historically, the approaches to fracture mechanics when significant plasticity has
occurred have been to consider either the crack tip opening displacement or the
J-integral. The current British Standard on fracture mechanics toughness tests, BS
7448: Part 1: 1991, describes how a single approach to linear elastic and general yielding fracture mechanics tests can be adopted.4
Significant yielding at a crack tip leads to the physical separation of the surfaces
of a crack, and the magnitude of this separation is termed the crack tip opening displacement (CTOD), and has been given the symbol d. The CTOD approach enables
critical toughness test measurements to be made in terms of dc, and then applied to
determine allowable defect sizes for structural components.
The J-integral is a mathematical expression that may be used to characterize the
local stress and strain fields around a crack front. Like the CTOD, the J-integral
simplifies to be consistent with the stress intensity factor approach when the conditions for linear elastic fracture mechanics prevail. When non-linear conditions
dominate either CTOD or J are useful parameters for characterizing the crack tip
fields.
The relationships between KI, d and JI under linear elastic conditions are
J I = GI =
d=
K I2
s YE
K I2
(1 - v 2 ) in plane strain
E
(7.4)
(7.5)
where GI is the strain energy release rate, which was the original, energy based
approach to studying fracture.
In all cases, either linear elastic or general yielding, there is a parameter that
describes the loading state of a cracked body. This might be KI, d or JI as appropriate to the conditions. The limiting case for a particular material is the critical value
at which failure occurs. Under elastic conditions this is the sudden fracture event,
and the material property is KIc. Under elastic–plastic conditions, failure is not
usually rapid, and the critical condition, dc or JIc, is usually associated with the onset
of the ductile fracture process.
The assessment of flaws in fusion welded structures is covered by BS 7910: 1999,9
which allows the use of material fracture toughness values in the form of KIc, dc or
JIc. In the absence of genuine fracture mechanics derived toughness data, estimates
of KIc may be made from empirical correlations with Charpy V-notch impact energies. Caution should be exercised when doing so as the degree of fit of the correlations tends to be poor.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
256
Fracture and fatigue
The BS 7910: 1999 document includes design curves for assessing the acceptability of a known flaw at a given stress level. There are three levels of sophistication
in the analysis requiring more precise information about the stresses and material
properties as the assessment becomes more advanced. The procedure is very powerful as it considers the possibility of either fracture or plastic collapse as alternative failure processes. The first level uses a simple approach with built-in safety
factors and conservative estimates of the material fracture toughness and the
applied and residual stresses in the structure. The standard allows for toughness estimates from Charpy impact tests to be used at Level 1. Level 2 is the normal assessment route for steel structures and requires more accurate estimates of the stresses,
material properties and defect sizes and shapes. Level 3 is much more sophisticated
and can accommodate the tearing behaviour of ductile metals. The details of dealing
with multiple flaws, residual stresses and combinations of bending and membrane
stresses are all dealt with in the document. In many practical cases, a Level 1 assessment is sufficient.
Annex D in BS 7910: 1999 describes the manual procedure for determining the
acceptability of a flaw in structure using the Level 1 procedure. An equivalent flaw
parameter, a , is defined as the half length of a through-thickness flaw in an infinite
plate subjected to a remote tension loading. An equivalent tolerable flaw parameter, a m, can then be estimated and used to represent a variety of different defect
shapes and sizes of equivalent severity:
am =
1 Ê Kmat ˆ
2p Ë s max ¯
2
(7.6)
Equivalent part-thickness flaw dimensions can then be estimated from graphical
solutions presented in Annex D of the standard. The possibility of plastic collapse
of the cracked sectioned must be checked by calculating the ratio, known as Sr, of
a reference stress to the flow stress of the material. The flow stress is taken to be
the average of the yield and tensile strengths. The reference stress is related to the
applied and residual stresses in the structure and depends on the geometry of the
structure and the defect. Details of its calculation are given in Annex C. Provided
that the Sr parameter is less than 0.8, there is no risk of failure by collapse and any
failure will be as a result of fracture.
The parameter Kmat in Equation (7.6) is a measure of the material fracture toughness, which may well not be valid plane strain KIc value but is appropriate to the
conditions and section sizes under review. Furthermore, Level 1 allows for the
possibility of estimating a fracture toughness value from Charpy impact data as
described in Annex E of the standard.
Tough structural steels are extremely tolerant of the presence of cracks. If one
considers a typical structural steel, with a yield stress of 275 N.mm-2 and a minimum
fracture toughness of 70 MPa÷m, subjected to a maximum allowable stress of
165 N.mm-2, then the maximum allowable flaw from Equation (7.6) is about 60 mm
long. (Remember that a m is the half length of a through-thickness crack in an infinite plate.) The correction for a finite width plate makes less than 10% difference
Steel Designers' Manual - 6th Edition (2003)
Materials testing for fracture properties
257
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
to the crack size provided that the width of the plate is more than four times the
length of the crack.
Reducing the maximum allowable stress in the structure has a big effect on the
allowable flaw size. Halving the maximum stress increases the allowable flaw size
to around 240 mm, a four-fold increase. The corollary, of course, is that the use of
high strength metals makes a structure more sensitive to the presence of defects.
High strength materials mean that higher structural stresses are allowed; high
strength steels also tend to have lower fracture toughnesses than low strength steels.
This combination means that the maximum allowable flaw size can become quite
small.
7.4 Materials testing for fracture properties
There are, essentially, two approaches to fracture testing: the traditional impact test
on a small notched bar, following the work of Izod or Charpy, or a fracture toughness test on a pre-cracked beam or compact tension specimen.
The principal advantage of an impact test is that it is relatively quick, simple and
cheap to carry out. It provides qualitative information about the relative toughness
of different grades of material and is well suited to quality control and material
acceptance purposes.
Fracture toughness tests on pre-cracked specimens provide a direct measure of
the fracture mechanics toughness parameters KIc, JIc or critical crack tip opening
displacement. They are, however, more expensive to perform, use larger specimens
and require more complex test facilities. The main advantage of fracture toughness
tests is that they provide quantitative data for the design and assessment of
structures.
7.4.1 Charpy test
Reference 1 specifies the procedure for the Charpy V-notch impact test. The test
consists of measuring the energy absorbed in breaking a notched bar specimen by
one blow from a pendulum as shown in Fig. 7.6. The test can be carried out at a
range of temperatures to determine the transition between ductile and brittle
behaviour for the material. The Charpy impact value is usually denoted by the
symbol Cv and is measured in joules.
Many attempts have been made to correlate Charpy impact energies with fracture toughness values. The large scatter found in impact data makes such relationships difficult to describe with any great degree of confidence. No single method is
currently able to describe the entire temperature–toughness response of structural
ferritic steels. The correlation method that forms the basis of Annex E in BS 7910:
1999 and also the latest European guidelines on flaw assessment methods10 is that
Steel Designers' Manual - 6th Edition (2003)
258
Fracture and fatigue
'T18 mm
J. max,
20
Iio mm
II
10 mm
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(0)
40 mm
between supports
V
8mmT
10 mm
mm
T
Iio mm
fcentreline
55 mm
(b)
Fig. 7.6
root
radius
0.25 mm
Charpy test. (a) Test arrangement; (b) specimen
known as the Master Curve. The temperature at which a certain specified Charpy
impact energy is achieved, usually 27J, is used as a fixed point, and the shape of the
transition curve is generated from that temperature. This is also the principle behind
the allowable section thickness guideline in BS 5400-3: 2000 and BS 5950: 2000.
7.4.2 Fracture mechanics testing
The recommended methods for determining fracture toughness values of metallic
material are described in BS 7448: Part 1: 1991.4 This covers both linear elastic and
elastic–plastic conditions. The advantage of a single test procedure is that the results
may be re-analysed to give a critical CTOD or critical J value if the test is found
not to conform to the requirements for a valid plane strain KIc result.
The principle behind the tests is that a single edge-notched bend or compact
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Materials testing for fracture properties
259
Fig. 7.7 Typical fracture toughness test specimens. (a) Three-point bend beam; (b) compact
tension specimen
tension specimen (see Fig. 7.7) is cyclically loaded within prescribed limits until a
sharp fatigue crack is formed. The specimen is then subjected to a displacementcontrolled monotonic loading until either brittle fracture occurs or a prescribed
maximum force is reached. The applied force is plotted against displacement and,
provided specific validity criteria are met, a plane strain KIc may be found by analysis of the data. When the validity criteria are not met, the data may be re-analysed
to evaluate a critical CTOD or critical J for that material. The determination of
critical CTOD requires the relationship between applied load and the opening of
the mouth of the crack, measured using a clip gauge. Critical J calculations need the
load against load line displacement response, so the detailed arrangements of the
two types of test are slightly different.
The specimen size requirements for a valid KIc are that:
Ê K Ic ˆ
thickness, width and ligament ≥ 2.5Á
˜
Ë sy ¯
2
(7.6)
The size requirements for the J value to be valid are that:
Ê J Ic ˆ
thickness, width and ligament ≥ 25Á ˜
Ësy ¯
(7.7)
This implies that significantly smaller specimens are required for critical J value tests
than for KIc tests.
Steel Designers' Manual - 6th Edition (2003)
260
Fracture and fatigue
7.4.3 Other tests
There are other tests that can be carried out such as the wide plate test used for
testing welded plate joints, which was developed by Wells at the Welding Institute.11
A large full-thickness plate, typically 1 m square, is butt welded using the process
and treatments to be used in the production weld. This test has the advantage of
representing failure of an actual welded joint without the need for machining prior
to testing.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
7.4.4 Test specimens
As has already been described, each test procedure requires a sample of a certain
size. In addition, the position of a sample in relation to a weld or, in the case of a
thick plate, its position through the thickness is important. In modern structural
steels toughness of the parent plate is rarely a problem. However, once a weld is
deposited the toughness of the plate surrounding the weld, particularly in the heataffected zone (HAZ), will be reduced. Although lower than for the parent plate, the
Cv or fracture toughness values that can be obtained should still provide adequate
toughness at all standard operating temperatures. In the case of thicker joints,
appropriate post-weld heat treatments should be carried out.
7.5 Fracture-safe design
The design requirements for steel structures in which brittle fracture is a consideration are given in most structural codes. There are several key factors which need
to be considered when determining the risk of brittle fracture in a structure. These
are:
(1)
(2)
(3)
(4)
minimum operating temperature
loading – in particular, rate of loading
metallurgical features such as parent plate, weld metal or HAZ
thickness of material to be used.
Each of these factors influences the likelihood of brittle fracture occurring.
From experimental data and parametric studies examining the maximum tolerable defect sizes in steels under various operating conditions, codes such as BS 54003 and BS 5950-1 provide tables giving the maximum permissible thickness of steel
at given operating temperatures. Where a design does not fit into these broad categories attempts have been made to use fracture mechanics to provide a criterion
for material selection in terms of Charpy test energy absorption as this information
is usually provided by the steel makers. When true fracture mechanics toughness
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fatigue
261
values are available, methods such as those described in References 8 and 9 should
be employed to assess the acceptability of flaws in fusion welded structures.
As discussed earlier, at normal operating temperatures and slow rates of loading
valid KIc values are not usually obtained for structural steels, and in the offshore
and nuclear industries critical CTOD and J tests are widely used. Under these conditions, valid KIc values from low temperature tests will be conservative. If the structure is acceptable when assessed using such conservative data then there is often no
great need to pursue the problem, particularly as all forms of fracture mechanics
testing are expensive compared with routine quality control tests such as Charpy
testing.
In general, the fracture toughness of structural steel increases with increasing
temperature and decreasing loading rates. The effect of temperature on fracture
toughness is well known. The effect of loading rate may be equally important, not
only in designing new structures, but also in understanding the behaviour of existing ones which may have been built from material with low toughness at their
service temperature. The shift in the ductile–brittle transition temperature for structural steels can be considerable when comparing loading rates used in slow bend
tests with those in Charpy tests. Results from experimental work have shown that
the transition temperature for a BS EN 10025 S355 J steel can change from around
0°C to -60°C with decreasing loading rate.
Materials standards set limits for the transition temperatures of various steel
grades based on Charpy tests. When selecting an appropriate steel for a given structure it must be remembered that the Charpy values noted in the standards apply to
parent plate. Material toughness varies in the weld and heat-affected zones of
welded joints, and these should be checked for adequate toughness. Furthermore,
since larger defects may be present in the weld area than the parent plate, appropriate procedures should be adopted to ensure that a welded structure will perform
as designed. These could involve non-destructive testing including visual examination. In situations where defects are found, fracture mechanics procedures such as
those in References 8 and 9 can be used to assess their significance.
7.6 Fatigue
7.6.1 Introduction
A component or structure which survives a single application of load may fracture
if the application is repeated a large number of times. This would be classed as
fatigue failure. Fatigue life can be defined as the number of cycles and hence the
time taken to reach a pre-defined failure criterion. Fatigue failure is by no means a
rigorous science and the idealizations and approximations inherent in it prevent the
calculation of an absolute fatigue life for even the simplest structure.
In the analysis of a structure for fatigue there are three main areas of difficulty
in prediction:
Steel Designers' Manual - 6th Edition (2003)
262
Fracture and fatigue
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(1) The operational environment of a structure and the relationship between the
environment and the actual forces on it;
(2) The internal stresses at a critical point in the structure induced by external
forces acting on the structure;
(3) The time to failure due to the accumulated stress history at the critical point.
There are three approaches to the assessment of fatigue life of structural components. The traditional method, called the S–N approach, was first used in the mid
19th century. This relies on empirically derived relationships between applied elastic
stress ranges and fatigue life. A development of the S–N approach is the strain–life
method in which the plastic strains are considered important. Empirical relationships are derived between strain range and fatigue life. The third method, based on
fracture mechanics, considers the growth rate of an existing defect. The concept of
defect tolerance follows directly from fracture mechanics assessments.
7.6.2 Loadings for fatigue
Fluctuating loads arise from a wide range of sources. Some are intentional, such
as road and rail traffic over bridges. Others are unavoidable, such as wave loading
on offshore oil rigs, and some are accidental – a car striking a kerb, for example.
Occasionally the loads are entirely unforeseen: resonance of a slender tower under
gusting wind loads can induce large numbers of small amplitude loads. References
12–14 are useful sources of information on fatigue loading.
The designer’s objective is to anticipate the sequence of service loading throughout the life of the structure. The magnitude of the peak load, which is vital for static
design purposes, is generally of little concern as it represents only one cycle in
millions. For example, highway bridge girders may experience 100 million significant cycles in their lifetime. The sequence is important because it affects the stress
range, particularly if the structure is loaded by more than one independent load
system.
For convenience, loadings are usually simplified into a load spectrum, which
defines a series of bands of constant load levels and the number of times that each
band is experienced, as shown in Fig. 7.8.
7.6.3 The nature of fatigue
Materials subject to a cyclically variable stress of a sufficient magnitude change
their mechanical properties. In practice a very high percentage of all engineering
failures are due to fatigue. Most of these failures can be attributed to poor design
or manufacture.
Fatigue failure is a process of crack propagation due to the highly localized cyclic
plasticity that occurs at the tip of a crack or metallurgical flaw. It is not a single
Steel Designers' Manual - 6th Edition (2003)
Fatigue
263
load
L
w2
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
fl
fl2
fl3
114
number of cycles
Fig. 7.8
Typical load spectrum for design
mechanism but the result of several mechanisms operating in sequence during the
life of a structure: propagation of a defect within the microstructure of the material; slow incremental propagation of a long crack; and final unstable fracture.
Crack initiation is a convenient term to cover the early stages of crack growth
that are difficult to detect. The reality is that a crack starts to grow from the first
loading cycles and continues right through to failure. In welded structures or cast
components the initiation phase is bypassed as substantial existing defects are
already likely to be present.
7.6.4 S–N curves
The traditional form for presenting fatigue data is the S–N curve, where the total
cyclic stress range (S) is plotted against the number of cycles to failure (N). A typical
curve is shown in Fig. 7.9 with a description of the usual terminology used in fatigue.
Logarithmic scales are conventionally used for both axes. However, this is not
universal, and S–N data may also be presented on linear stress axes instead of
logarithmic; stress amplitudes instead of ranges and reversals instead of cycles.
Fatigue endurance data are obtained experimentally. S–N curves can be obtained
for a material, using smooth laboratory specimens, for components or for detailed
sub-assemblies such as welded joints. In all cases, a series of specimens is subject
to cycles of constant load amplitude to failure. A sufficient number of specimens
are tested for statistical analysis to be carried out to determine both mean fatigue
strength and its standard deviation. Depending on the design philosophy adopted,
design strength is taken as mean minus an appropriate number of standard
deviations.
Steel Designers' Manual - 6th Edition (2003)
264
Fracture and fatigue
R = S/S....
t alternating stress
maximum
stress Sm.
__ ___ ___
a
stress
mean
range S,
I
stress S,
.i Trninirnum
I
I
stress Sr.r
time
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
1000
100 ndurandurance limit
10
i0
i0
10
iO
10
10'
design life N (cycles)
Fig. 7.9
Typical S–N curve and nomenclature used in fatigue
Under some circumstances, laboratory tests on steel specimens appear to have
an infinite life below a certain stress range. This stress range is variously known
as the fatigue limit or endurance limit. In practice, the tests are stopped after
two, ten or one hundred million cycles, and if it has not broken then the stress
range is assumed to be below the fatigue limit. The fatigue, or endurance, limit
will tend to disappear under variable-amplitude loading or in the presence of a
corrosive environment. This means that real components will eventually fail
whatever the stress range of the loading cycles, but the life may be very long
indeed.
Welded steel joints are usually regarded as containing small defects due to
the welding process itself. It has been found after much experimental work
that the relationship between fatigue life and applied stress range follows the
form
NfDs a = b
(7.8)
Steel Designers' Manual - 6th Edition (2003)
Fatigue
265
1000
100
S
B
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
C
D
E
10
100,000
Fig. 7.10
1,000,000
10,000,000
Endurance, N (cycles)
100,000,000
Fatigue design curves for welded joints according to BS 5400-10: 1980 and BS 7608:
1993
where Nf is the number of cycles to failure, Ds is the applied stress range, and a and
b are constants which depend on the geometry of the joint. The value of a is in the
range of 3 to 4 for welded joints in ferritic steels.
The British Standard BS 7608: 1993 contains a code of practice for fatigue design
of welded structures which is consistent with the assessments for fatigue in steel
bridges in BS 5400-10: 19808 and other welded structures in BS 7910: 1999.9 The
basis is a series of design S–N curves for different welded joint configurations
and crack locations. Each potential crack location in each type of joint is classified
by a letter: S, B, C, D, E, F, F2, G or W. Each detail class has a specific design
S–N curve (Fig. 7.10). This curve indicates the number of cycles at any given stress
range that should be achieved with 97.7% probability of surviving. The design S–N
curves for each class of joint are based on extensive experimental data and are suitable for ferritic steels. Details of dealing with other metals are given in BS 7910:
1999.
The procedure is to identify the worst weld detail in the design and the ranges
of the stress cycles experienced by the structure. If there is only one stress range,
then the corresponding lifetime can be read off from the appropriate S–N curve.
This is the number of repeated cycles that the structure can endure and have a
97.7% probability of survival.
Consider the weld detail in Fig. 7.11. This is classified as type F2. If this is subjected to a cyclic stress with a range of 20 N.mm-2 every 7 seconds, then reading off
the type F2 curve in Fig. 7.10 indicates that the structure should survive 66 million
cycles. This corresponds to 4.8 ¥ 108 seconds or about 14 years.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
266
Fig. 7.11
Fracture and fatigue
An example of an F2 cracked weld
7.6.5 Variable-amplitude loading
For constant-amplitude loading, the permissible stress range can be obtained
directly from Fig. 7.10 by considering the required design life. In practice it is more
common for structures to be subjected to a loading spectrum of varying amplitudes
or random vibrations. In such cases use is made of Miner’s rule.15
Miner’s rule is a linear summation of the fatigue damage accumulated during the
life of the structure. For a joint subjected to a number of repetitions, ni, each of
several stress ranges Dsi, the value of ni corresponding to each Dsi should be determined from stress spectra measured on similar equipment or by making reasonable
assumptions as to the expected service history. The permissible number of cycles,
Ni, at each stress range, Dsi, should then be determined from Fig. 7.10 for the relevant joint class and the stress range adjusted so that the linear cumulative damage
summation does not exceed unity:
i= j
nj
n1
n2
n3
ni
+
+
+ ...+
=Â
< 1.0
N1 N 2 N 3
N j i =1 N i
(7.9)
The order in which the variable-amplitude stress ranges occur in a structure is not
considered in this procedure.
An example would be a class F2 weld that is subjected to 3 ¥ 107 cycles at a stress
range of 20 N.mm-2 and then the stress range increases to 30 N.mm-2 and the remaining life needs to be estimated. The lifetime at 20 N.mm-2 is read from Fig. 7.10 and
Steel Designers' Manual - 6th Edition (2003)
Fatigue
267
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
is 6.6 ¥ 107 cycles. The fraction of life used is therefore (3 ¥ 107)/(6.6 ¥ 107) = 0.455
and the remaining life fraction is 0.545. At a stress range of 30 N.mm-2 the lifetime
from Fig. 7.10 is 1.9 ¥ 107 cycles, so the remaining life for this welded joint is
0.545 ¥ 1.9 ¥ 107 = 1 ¥ 107 cycles.
Various methods exist to sum the spectrum of stress cycles. The rainflow counting method is probably the most widely used for analysing long stress histories using
a computer. This method separates out the small cycles that are often superimposed
on larger cycles, ensuring both are counted. The procedure involves the simulation
of a time history, or use of measure sequences, with appropriate counting algorithms.
Once the spectrum of stress cycles has been determined, the load sequence is broken
down into a number of constant load range segments. The reservoir method, which
is easy to use by hand for short stress histories, is described in Reference 9.
7.6.6 Strain–life
The notion that fatigue is associated with plastic strains led to the strain–life
approach to fatigue. The endurance of laboratory specimens is correlated to plastic
strain range, or amplitude, in a strain-controlled fatigue test (Fig. 7.12). A common
empirical curve fit to the fatigue lifetime data takes the form:
ea =
s ¢f
b
c
(2 N f ) + ef¢(2 N f )
E
(7.10)
•1
'0
0.01
-
— - AISI 4340, Tensile strength =1170 MPa
-
L—A1S1 4340, Tensile strength =1470 MPa
,,.,...,I
0.001
100
1,000
10,000
100,000
1,000,000
Cycles to failure
Fig. 7.12
Strain–life curves for a high strength steel in two conditions
10,000,000
Steel Designers' Manual - 6th Edition (2003)
268
Fracture and fatigue
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
where E is Young’s modulus and s¢f, e¢f, b and c are considered to be material properties. It is worth noting that the first term of the right hand side is an elastic stress
term which dominates at low loads and long lifetimes. The second term is a plastic
strain term which dominates at high loads and short lifetimes.
The strain–life approach is preferred over the S–N method since it is almost identical to the S–N approach at long lives and elastic stresses, and is more general for
problems of short lives, high strains, high temperatures or localized plasticity at
notches.
The technique is:
(1) determine the strains in the structure, often by finite element analysis;
(2) identify the maximum local strain range, the ‘hot spot’ strain;
(3) read, from the strain–life curve, the lifetime to first appearance of a crack at
that strain at that position.
Variable-amplitude loads are dealt with in the same way as in the S–N method, with
the local hot spot strains and their associated lifetimes being determined for each
block of loading.
The strain–life, or local strain, method has wider use in fatigue assessments in the
engineering industry than the S–N method and is available in commercial computer
software.
7.6.7 Fracture mechanics analysis
Fatigue life assessment using fracture mechanics is based on the observed relationship between the change in the stress intensity factor, DK, and the rate of growth of
fatigue cracks, da/dN. If experimental data for crack growth rates are plotted against
DK on a logarithmic scale, an approximate sigmoidal curve results, as shown in Fig.
7.13. Below a threshold stress intensity factor range, DKth, no growth occurs. For
intermediate values of DK, the growth rate is idealized by a straight line. This
approach was first formulated by Paris and Erdogan,16 who proposed a power law
relation of the form:
da
m
= A(DK )
dN
(7.11)
where da/dN is the crack extension per cycle, A, m are crack growth constants, and
DK = Kmax - Kmin
where Kmax and Kmin are the maximum and minimum stress intensities respectively
in each cycle. Since the crack growth rate is related to DK raised to an exponent,
it is important that DK should be known accurately if meaningful crack growth
predictions are to be made.
Steel Designers' Manual - 6th Edition (2003)
Fatigue
269
KIc
—3
non—continuum
mechanisms
continuum
mechanism
final
failure
(striation growth)
/
0
e
0
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
•0
/
dA/dN=Am
—6
B
m
static failure
mechanisms
—9
log10 uK
Fig. 7.13
Schematic presentation of crack growth
Values of A and m to describe the fatigue crack growth rate can be obtained from
specific tests on the materials under consideration. From published data reasonable
estimates of fatigue growth rates in ferritic steels are given by9
m=3
A = 5.21 ¥ 10-13 for non-aggressive environments at temperatures up to 100°C
A = 2.3 ¥ 10-12 for marine environments at temperatures up to 20°C
for crack growth rates in mm/cycle and stress intensity factors in N.mm-1.5.
Some care should be taken with fatigue lifetime estimates in corrosive environments to make sure that the fatigue crack growth data are appropriate to the particular combination of steel grade and environment. Small changes in steel
composition or environmental conditions can result in very large changes in crack
growth behaviour.
The fracture mechanics convention in the welded steel structures industry is
slightly different from that of other practitioners. For a crack at the toe of a welded
joint
DK = MK YDs pa
where Ds = applied stress range
a = crack depth
Y = a correction factor dependent on crack size, shape and loading
(7.12)
Steel Designers' Manual - 6th Edition (2003)
Fracture and fatigue
270
MK = a function which allows for the stress concentration effect of the joint
and depends on crack site, plate thickness, joint and loading.
In many industries the magnification factor, MK, is incorporated in the geometric
correction term, Y.This makes Equation (7.12) the fatigue version of Equation (7.1).
Substituting Equation (7.12) in Equation (7.11), rearranging and integrating gives
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Nf =
1
Ap
m 2
Ds
m
Ú
af
ai
da
[ MK Y
m
a]
(7.13)
where ai is the initial crack depth and af is the final crack depth corresponding to
failure.
Equation (7.13) forms an extremely powerful tool. If a welded joint contains a
crack or crack-like flaw, then Equation (7.13) can be used to predict its fatigue
endurance. This makes the reasonable assumption that the life consists of crack
growth from a pre-existing crack. The techniques requires knowledge of:
(1)
(2)
(3)
(4)
the crack propagation behaviour described by Equation (7. 11),
the initial flaw size,
the final flaw size,
the geometry and loading correction terms Y and MK.
The integration is straightforward if Y and MK are independent of crack length,
which is not usually the case. Otherwise some suitable numerical technique must be
used. The life is fairly insensitive to the final crack length, but highly dependent on
the initial flaw size. Engineering judgement is often required when selecting appropriate values of these crack sizes.
If Y and MK are independent of crack length, and m π 2, then
Nf =
Ê mˆ
Á 1- ˜ ˘
È ÊËÁ 1- m2 ˆ¯˜
ai
- af Ë 2 ¯ ˙
Í
m 2
Î
˚
(m 2 - 1) Ap (Mk YDs )
1
m
(7.14)
The initial crack size can be taken as the largest flaw that escapes the detection
technique used. This is likely to be several millimetres for visual or ultrasonic inspection of large welded structures.
The final allowable flaw size might be taken from Equation (7.5), or it may be a
crack that penetrates the wall of a containment vessel and allows fluid to escape.
Other failure conditions could be a crack that allows excessive displacements to
occur or a crack that is observable to the naked eye and is therefore unacceptable
to the customer.
The design S–N curves, described by Equation (7.8), and given in the British
Standards BS 5400-10: 1980, BS 7608: 1993 and BS 7910: 1999, are effectively the
same as Equation (7.14), the integration of the crack propagation equation.
Steel Designers' Manual - 6th Edition (2003)
Fatigue
271
7.6.8 Improvement techniques
7.6.8.1 Introduction
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
The fatigue performance of a joint can be enhanced by the use of weld improvement techniques. There is a large amount of data available on the influence of weld
improvement techniques on fatigue life but as yet little progress has been made
into developing practical design rules. Modern steelmaking has led to the production of structural steels with excellent weldability. The low fatigue strength of a
welded connection is generally attributed to a short crack initiation period. An
extended crack initiation life can be achieved by:
(1) reducing the stress concentration of the weld,
(2) removing crack-like defects at the weld toe,
(3) reducing tensile welding residual stresses or introducing compressive stresses.
The methods employed fall broadly into two categories:
weld geometry improvement: grinding: weld dressing; profile control
residual stress reduction: peening; thermal stress relief
Most of the current information relating to weld improvement has been obtained
from small-scale specimens. When considering actual structures one important
factor is size. In a large structure, long-range residual stresses due to the assembly
of the members are present and will influence the fatigue life. In contrast to small
joints, where peak stress is limited to the weld toe, the peak stress region in a large
multi-pass joint may include several weld beads, and cracks may initiate anywhere
in this highly stressed area.
However, it is good practice not to seek benefit from improvement techniques in
the design office.
7.6.8.2 Grinding
The improvement of the weld toe profile and the removal of slag inclusions can be
achieved by grinding either with a rotary burr or with a disc. To obtain the maximum
benefit from this type of treatment it is important to extend the grinding to a sufficient depth to remove all small undercuts and inclusions. The degree of improvement achieved increases with the amount of machining carried out and the care
taken by the operator to produce a smooth transition.
The performance of toe-ground cruciform specimens is fully investigated in
Reference 11. Under freely corroding conditions the benefit from grinding is
minimal. However, in air, the results appear to fall on the safe side of the mean
curve: endurance is altered by a factor of 2.2. It is therefore recommended that an
Steel Designers' Manual - 6th Edition (2003)
272
Fracture and fatigue
increase in fatigue life by a factor of 2.2 can be taken if controlled local machining
or grinding is carried out.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
7.6.8.3 Weld toe remelting
Weld toe remelting by TIG and plasma arc dressing are performed by remelting the
toe region with a torch held at an angle of 50° or 90° to the plate (without the addition of filler material). The difference between TIG and plasma dressing is that the
latter requires a higher heat input.
Weld toe remelting can result in large increases in fatigue strength due to the
effect of providing low contact angle in the transition area between the weld and
the plate and by the removal of slag inclusions and undercuts at the toe.
7.6.8.4 Hammer peening
Improved fatigue properties of peened welds are obtained by extensive cold
working of the toe region. These improved fatigue properties are due to:
(1) introduction of high compressive residual stresses,
(2) a flattening of crack-like defects at the toe,
(3) an improved toe profile.
It can be shown that weld improvement techniques greatly improve the fatigue life
of weldments. For weldments subject to bending and axial loading, peening appears
to offer the greatest improvement in fatigue life, followed by grinding and TIG
dressing.
7.6.9 Fatigue-resistant design
The nature of fatigue is well understood and analytical tools are available to calculate the fatigue life of complex structures. The accuracy of any fatigue life calculation is highly dependent on a good understanding of the expected loading sequence
during the whole life of a structure. Once a global pattern has been developed then
a more detailed inspection of particular areas of a structure, where the effects of
loading may be more important, due to the geometries of joints for example, should
be carried out.
Data have been gathered for many years on the performance of bridges, towers
cranes and offshore structures where fatigue is a major design consideration. Codes
of practice, such as BS 5400-10 and BS 7608, give details for the estimation of fatigue
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
References
273
lives. Where a structure is subjected to fatigue it is important that welded joints are
considered carefully. Fatigue and brittle fractures can be initiated at discontinuities
of shape, at notches and cracks which give rise to high local stress. Avoidance of
local structural and notch peak stresses by good design is the most effective means
of increasing fatigue life. It is important that during the design process consideration is given to the manner in which the structure is to be fabricated. Acute-angled
welds of less than 30° are difficult to fabricate, particularly in tubular structures. This
could lead to defects in the weld. Furthermore it is also difficult to carry out nondestructive testing on such welds. Despite these problems, some recently designed
structures have incorporated such features.
Repairs carried out to structures in service are expensive, and in the worst case
may require that a facility is closed down temporarily. Care needs to be taken when
specifying secondary attachments to main primary steelwork. These are often not
considered in detail during the design process as they themselves are not complex.
However, there have been a number of failures in offshore structures due to fatigue
crack growth resulting from a welded attachment. The following general suggestions
can assist in the development of an appropriate design of a welded structure with
respect to fatigue strength:
(1) Adopt butt or single and double bevel butt welds in preference to fillet
welds.
(2) Use double-sided in preference to single-sided fillet welds.
(3) Aim to place weld, particularly toe, root and weld end in area of low stress.
(4) Ensure good welding procedures are adopted and adequate non-destructive
testing (NDT) undertaken.
(5) Consider the effects of localized stress concentration factors.
(6) Consider potential effects of residual stresses.
References to Chapter 7
1. British Standards Institution (1990) Charpy impact test on metallic materials.
BS EN 10045-1, BSI, London.
2. British Standards Institution (1990) Hot rolled products of non-alloy structural
steels, BS EN 10025, BSI, London.
3. American Society for Testing and Materials (1981) The standard test for JIc a
measure of fracture toughness. ASTM E813-81.
4. British Standards Institution (1991) Fracture mechanics toughness tests. Part 1.
Method for determination of KIc, critical CTOD and critical J values of metallic
materials. BS 7448-1: 1991, BSI, London.
5. Murakami Y. (1987) Stress Intensity Factors Handbook. Pergamon Press,
Oxford.
6. Tada H., Paris P.C. & Irwin G.R. (1985) The Stress of Cracks Handbook. Del
Research Corporation, Hellertown, Pa., USA.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
274
Fracture and fatigue
7. British Standards Institution (2000) Structural use of steelwork in building. Code
of practice for design. BS 5950-1: 2000.
8. British Standards Institution (2000) Steel, concrete and composite bridges. Code
of practice for design of steel bridges. BS 5400-3: 2000 and Code of practice for
fatigue. BS 5400-10: 1980, BSI, London.
9. British Standards Institution (1999) Guidance on methods for assessing the
acceptability of flaws in fusion welded structures. BS 7910: 1999, BSI, London.
10. SINTAP (1999) Structural Integrity Assessment Procedures for European
Industry. Project BE 95-1426. Final Procedure, British Steel Report, Rotherham,
UK.
11. American Society for Testing and Materials (1982) Design of fatigue and fracture resistant structures. ASTM STP 761.
12. British Standards Institution (1993) Fatigue design and assessment of steel structures. BS 7608: 1993, BSI, London.
13. Department of Energy (1990) Offshore Installations: Guidance Design
Construction and Certification, 4th edn. HMSO.
14. British Standards Institution (1983 & 1980) Rules for the design of cranes.
Part 1: Specification for classification, stress calculations and design criteria for
structures. Part 2: Specification for classification, stress calculations and design of
mechanisms. BS 2573, BSI, London.
15. Miner, M.A. (1945) Cumulative damage in fatigue. Journal of Applied
Mechanics, 12, A159–A164.
16. Paris P.C. & Erdogan F. (1963) A critical analysis of crack propagation laws.
Journal of Basic Engineering, 85, 528–534.
Further reading for Chapter 7
Dowling N.E. (1999) Mechanical Behavior of Materials: Engineering Methods for
Deformation, Fracture, and Fatigue, 2nd edn. Prentice-Hall, Inc.
Gray T.F.G., Spence J. & North T.H. (1982) Rational Welding Design, 2nd edn.
Butterworth, London.
Gurney T.R. (1979) Fatigue of Welded Structures, 2nd edn. Cambridge University
Press.
Pellini W.S. (1983) Guidelines for Fracture-Safe and Fatigue-Reliable Design of Steel
Structures. The Welding Institute.
Radaj D. (1990) Design and Analysis of Fatigue Resistant Welded Structures.
Abington Publishing, Cambridge.
Full Page Ad, 3mm Bleed
Steel Designers' Manual - 6th Edition (2003)
Chapter 8
Sustainability and
steel construction
by GRAHAM RAVEN
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
8.1 Introduction
The UK Government, along with many others, has a very strong policy to encourage sustainable development. In May 1999, it published A better quality of life – a
strategy for sustainable development for the United Kingdom. The principles in this
document were further developed for the construction industry in Building a better
quality of life – a strategy for more sustainable construction, which was published in
April 2000. The underlying principles of sustainability lie in the appropriate balance
of economic, social and environmental impacts, the so-called triple bottom line. The
strong message is that for a viable long term future any enterprise must pay due
concern to the following issues:
•
•
•
•
Maintenance of high and stable levels of economic growth and employment
Social progress which recognizes the needs of everyone
Prudent use of natural resources
Effective protection of the environment.
Clearly, to be successful all these issues need to be balanced, as over-commitment
to any one of them at the expense of the others will lead to failure in the long term.
There are many examples of how the steel construction sector has progressed to
be itself more sustainable and contribute to sustainable construction. Some of these
are set out below.
Steel has a significant role in construction and has established a strong position.
In the UK structural frames now have approximately 70% of the multi-storey frame
and 95% of the single storey industrial buildings market.
8.2 Economic impacts
The UK steelwork sector has worked and continues to work hard to provide the
most economic solutions and work methods. It has become world-class through its
investments in IT and associated technologies, based on the use of 3-D modelling
for detailing with direct links to CNC machines and suppliers. This is proving to be
a strong foundation for the release of further benefits which stems from integration
275
Steel Designers' Manual - 6th Edition (2003)
276
Sustainability and steel construction
with design activities through the adoption of data transfer protocols such as the
CIMSteel Integration Standards.
Since long before the needs of sustainability were fully recognized, there has been
a history of continuous improvement in structural design methods which have
resulted in economies to clients over a period of years. This has been encouraged
and has been complemented by the development of systems and components, many
involving the use of light gauge steel in addition to the more traditional hot rolled
sections.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
8.3 Social impacts
For any enterprise to be successful over a reasonable period it needs skilled and
conscientious people. The steel construction sector continues to invest heavily in the
education and training of both its employees and its customers in the efficient and
safe use of its products. Considerable importance in the sustainability agenda is
attached to the effects of enterprises on the local community. A factory production
environment which encourages a stable workforce is more conducive to both
employers and employees in the promotion of skills development. It is obviously
easier for a relationship with the local community to develop in such an environment, than where there is a predominantly casual or itinerant workforce.
On site, fast dry construction from off-site manufactured assemblies is less disruptive to neighbours and provides earlier weatherproofing and hence more reliable and acceptable working conditions for other trades. This contributes to safer
working and a higher quality product. The risk of water pollution from wet trades
is also minimized.
8.4 Environmental impacts
8.4.1 Effective protection of the environment
One of the key drivers for change is the need to reduce global warming gas emissions. This is largely achieved through conservation of both embodied and operational non-renewable energy. Operational energy is the most significant, and SCI is
active in several areas. Research has shown that adaptable solutions can be designed
using composite construction, which, even with its efficient use of material, still provides sufficient thermal capacity for fabric energy storage systems to be effective.1
This work is being extended to take advantage of air and water cooling of floor
systems. Of course precast concrete floors can also be most efficiently incorporated
into steel framed buildings, as demonstrated by the Wessex Water headquarters
building. This boasts one of the lowest energy consumptions in the UK for buildings of this type and utilizes a lightweight structural steel frame.2
Steel Designers' Manual - 6th Edition (2003)
Embodied energy
277
For the occupancy patterns of sheds and residential buildings, fabric energy
storage systems are less appropriate and high insulation values and air tightness are
more relevant. These can be conveniently achieved in steel-framed construction
and, in the case of sheds, through the use of well-detailed, insulated steel cladding
systems.
Prefabrication has a major contribution in minimizing pollution. All forms of
waste to landfill, as well as noise, dust, emissions and other potential contamination
on building sites, can be more easily controlled when most of the work is carried
out under factory conditions.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
8.4.2 Prudent use of natural resources
Originally driven by economic reasons, the history of design development has led
to methods which optimize the use of natural resources. UK steel design methods
are among the most efficient internationally. Resources are conserved not only by
minimizing them at first use but also by extending their lifetime. The life of steel
structures is extended through both the ability to create flexible spaces from the
use of long spans and the adaptability of the structure during its lifetime to accommodate change of usage. When decommissioning is necessary, the elements may
be re-used or, if this is not viable, recycled. Recent studies3 have shown that
at present 10% of steel arising from construction demolition is re-used, with a
further 85% recycled. Furthermore, it is practical to design structures in steel which
are suited to deconstruction rather than demolition and so can be re-used.
This further enhances the efficiency of the use of all the resources embodied in the
components.
In general, off-site factory production is easier to control and manage and so leads
to less waste of both materials and other resources. The steel construction sector is
well advanced in the production of modules which include services and other trades
as well as steel components.
This short review demonstrates that design for sustainability is an extremely
broad and important area, and steel can help provide solutions in a wide variety of
ways. The remainder of the chapter concentrates on more specific issues influencing the environmental impacts.
8.5 Embodied energy
Embodied energy is the energy consumed in the process of manufacturing, using
and later disposing of, or recycling, materials. Whilst a substantial amount of energy
is required to manufacture steel, it is a highly efficient structural material and relatively little energy is required for the rolling process to form structural sections.
Overall therefore steel structures tend to have good embodied energy profiles.
Furthermore, steel is a material that lends itself extremely well to the sustainability
hierarchy of ‘reduce, re-use and recycle’.
Steel Designers' Manual - 6th Edition (2003)
278
Sustainability and steel construction
8.5.1 Reduction
Steel is a material which lends itself to prefabrication. Structural components such
as beams and columns are delivered to site as finished items; wastage during the
construction process is therefore minimal. Continuous improvement in structural
design has led to British Standards that give design rules which result in optimum
use of material while maintaining high standards of safety.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
8.5.2 Re-use
Re-using materials requires less energy than recycling, although it is likely that
some refurbishment will be necessary. Steel structures are conceptually a kit of
parts; with appropriate design and detailing of connections both between steel
elements and other components there is considerable potential to increase the 10%
of structures currently re-used.3
The ability to dismantle steel structures on a piece-by-piece basis minimizes the
environmental problems often associated with demolition such as noise, dust and
safety hazards. Furthermore, steel foundation and substructure components can
usually be extracted and recovered at the end of a building’s life, eliminating ground
contamination. The high monetary value of steel compared with other construction
materials justifies the expense involved in recovery.
There are obviously barriers to overcome, for example assuring the next
user of the history and quality of the components and the need to provide the
necessary storage and accessibility to available components; however, these can be
overcome.
To maximize the potential for re-use and minimize any fabrication needed for
subsequent users a higher degree of standardization than is normal today is very
desirable. Since solutions are technically feasible, designers and suppliers are reacting to the needs and pressures of society to put the necessary processes into place.
8.5.3 Recycling
Steel is 100% recyclable, and repeated melting, casting and rolling have no
detrimental effect on quality. 50% of all steel production worldwide is from recycled
sources. This considerably reduces embodied energy, pollution and resource
depletion.
Scrap steel has a monetary value and there are established international markets
for steel scrap. The waste products generated from the production of steel such as
ferro-lime (steel slag) and blast-furnace slag are also recycled and used as substitutes
for primary aggregates and cement. As mentioned previously, recent studies as part
of a pan-European project into life cycle assessment of steel construction products
Steel Designers' Manual - 6th Edition (2003)
Embodied energy
279
have shown that currently over 85% of steel from demolition of buildings is recycled and a further 10% re-used.3
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
8.5.4 Housing and embodied energy
In a typical new house complying with current Building Regulations, the embodied
energy of construction may be equivalent to approximately 5 years of operational
energy. However, with increasing insulation and air-tightness standards, which
together are reducing operational energy requirements, the embodied energy is
becoming more significant.
Residential light steel frame construction is usually based on close centre frameworks made of cold-formed studs and joists (generally ‘C’ or ‘Sigma’ sections).These
sections, produced from galvanized steel by roll forming, are usually 1.2–3.2 mm
thick. Sections and small panels can often be manhandled into position; large panels
can be erected using small capacity cranes.
The inherent strength and flexibility of light steel framing allows the utilisation
of roof spaces as additional useable area. This reduces the footprint of dwellings
relative to their internal floor area, which is particularly advantageous for high
density low rise urban developments.
Although steel has a relatively high embodied energy content (approximately
18 MJ/kg), it is used efficiently in light steel frame construction. Approximately
20 kg/m2 of steel are used in a typical house frame, resulting in an overall embodied
energy content of 360 MJ/m2. Investigations suggest this is approximately equal to
the figure for traditional construction, with any differences due in general to the
particular circumstance rather than the choice of material.
Furthermore, the higher strength to weight ratio compared with alternative
materials allows the design of buildings with greater spans, fewer internal loadbearing walls and fewer foundations.
Light steel house frames can easily be strengthened, extended, modified, dismantled, reassembled and repaired, and so have a long service life.
Light steel modular construction offers further advantages. It reduces transport
requirements by around 50% compared with conventional construction. Building
materials are delivered in bulk to the factory and prefabricated modules delivered
to the site. The number of journeys to the site is therefore minimized, reducing noise
and nuisance as well as energy in transportation. In the case of volumetric modular
construction, whole rooms may be completed fitted out in the factory.
8.5.5 Non-residential buildings and embodied energy
The embodied energy burdens of commercial buildings vary considerably depending upon their specification, as illustrated in Fig. 8.1. For example, prestige office
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
280
Sustainability and steel construction
Fig. 8.1
Comparison of embedded energy and operating energy for a basic grade and
prestige office over a 60-year life
buildings can have twice the embodied energy content of lower specification offices.
This is due to the way they are clad and fitted out rather than the effects of the
structural frame.
Embodied energy is typically small in comparison to operational energy over the
life of a commercial building (generally an order of magnitude less). The balance is
however changing as improved designs reduce operational energy and the rate of
change of use increases so requiring more refurbishment.
The embodied energy of the structural components of a typical office building is
only about 25–30% of the total embodied energy and tends to vary relatively little
irrespective of the type of structure.A life cycle assessment (LCA)4 study of a typical
four storey office building compared five alternative steel and concrete structures.
No significant difference was identified between the embodied energy burdens of
the different options. Although steel has a high embodied energy content per kilogram, its high strength to weight ratio means that a much lower volume of steel is
required to construct a steel frame than the volume of concrete needed to construct
an equivalent frame. The cross-sectional area and mass of steel structural members
is very much smaller than that of equivalent concrete members.
The comparative lightness of modern steel frame structures has been important
in reducing their embodied energy burdens. In addition to savings associated with
the frame materials, designers can often adopt lighter foundation systems than
would be required for other forms of construction. Alternatively the number of
foundations may be reduced, since the spanning capabilities of steel can allow larger
structural bays, and consequently fewer column bases requiring support. The larger
bays also allow for more flexible use of space; steel frames are easily adaptable,
thereby prolonging the building life and reducing yet further the impact of embodied energy over the life cycle.
In composite construction, where beams act in tandem with the floor slab, the
additional strength and stiffness that are achieved allow the use of relatively shallow
beams. These have lower embodied energy and the volume of concrete in the floor
is barely altered.
Steel Designers' Manual - 6th Edition (2003)
Operational energy
281
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Modern portal frame design has played a significant role in reducing the embodied energy burden of single storey industrial and commercial buildings. Portal
frames are capable of spanning considerable distances using relatively light sections.
Roof beam depths are generally of the order of span divided by 60, as opposed to
span divided by 25 for a conventional beam and column roof. The combination of
steel frames with light gauge purlins and rails supporting insulated cladding systems
provides a very efficient form of construction resulting in the very high (95%) major
share. Systems are available to meet the requirements of the new Part L to the UK
Building Regulations5 in terms of both insulation and air tightness.
A key element in achieving low embodied energy is the pursuit of structural
efficiency. By reducing the amount of structural steel required per unit area
of commercial building, designers can not only achieve elegant and economic
structural solutions, but also reduce embodied energy burdens.
8.6 Operational energy
Buildings require energy for heating and lighting, powering equipment and machines, and possibly mechanical ventilation and cooling services. This type of energy,
known as operational energy, is greatly influenced by the design of the building, and
the servicing strategy.
The threat of global climate change resulting from ‘greenhouse gases’, of which
the most prevalent is carbon dioxide, is driving a concerted effort to reduce energy
use in buildings. In the UK, buildings account for 40–50% of carbon dioxide emissions. Significant reductions in operational energy (and hence CO2 emissions) are
necessary to combat climate change. Through careful design, it is possible to reduce
operational energy by at least one third, by reducing the need for air conditioning
or even eliminating it altogether.2
8.6.1 Housing and operational energy
Residential buildings account for about 30% of the total energy used in the UK.
This relatively high percentage of energy use, together with increasing demand for
new housing, indicates the importance of achieving improved energy efficiency in
this sector.
Light steel frame residential buildings have proved to have good standards of
environmental performance, particularly in terms of energy conservation. Many
have significantly exceeded the requirements of UK Building Regulations in terms
of thermal performance, as shown in Table 8.1.
The following measures are recommended to reduce operational energy
requirements:
Steel Designers' Manual - 6th Edition (2003)
282
Sustainability and steel construction
Table 8.1 Comparison of thermal performance of light steel frame residential buildings with UK
Building Regulations
U-values (W/m2K)
Walls
Floor
Roof
Windows
1995 Building Regulations
2002 Building Regulations
Good practice1
Ultra-low energy homes2
Low energy steel frame house3
0.45
0.35
0.35
0.20
0.20
0.45
0.25
0.32
0.20
0.35
0.25
0.20
0.20
0.15
0.20
3.3
2.2
2.8
2.0
1.8
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Notes:
1. Good Practice Guide 79 – Energy efficiency in housing, Department of Environment, 1993.
2. General Information Report 38 – Review of ultra-low energy homes, Best Practice Programme of the
Department of Environment, 1996.
3. U-values of the Oxford Brookes University Demonstration Building.
•
•
•
•
Minimization of heat loss through the building envelope
Provision of an airtight building envelope with controlled ventilation
Provision of efficient and controlled heating
Use of appropriate passive heating, cooling and ventilation methods.
These measures can often be achieved at little or no extra overall cost, particularly
where savings can be made in the capital cost of heating systems as a result of additional insulation and airtight construction. To accompany the regulations, a document is being developed showing robust details of plans which will, if properly
constructed, give compliance with the UK Building Regulations. To achieve compliance with the new regulations5 will be easier with framed construction than
with traditional methods.
Light steel frame housing has several generic advantages but as with all forms
of construction attention should be paid to the detailing in order to maximize the
benefits.
Placing the insulation on the outside of the frame is known as warm frame construction; this minimizes heat loss since there is minimal thermal bridging. Risk of
condensation is eliminated since all the members are contained within the warm
internal environment. High levels of insulation are possible however without
significantly increasing wall thickness by the addition of insulation between the
frame members. Heat loss resulting from air infiltration can also be conveniently
minimized by the inclusion of an air tight membrane in the walls.
8.6.2 Commercial buildings and operational energy
Commercial buildings use energy primarily for space and water heating, lighting,
cooling, ventilation and small power. In order to reduce operational energy, buildings should be designed to provide maximum occupant comfort throughout the year
with minimum energy requirements. The charts in Fig. 8.2 show typical energy
requirements and carbon dioxide emissions for two types of commercial building.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Operational energy
Fig. 8.2
283
Comparison of operational energy requirements and resulting carbon emissions for
a typical air conditioned office and a good practice naturally ventilated open plan
office (source: Econ 19, Energy use in offices, Energy Efficiency Best Practice
Programme)
Although heating accounts for a significant part of the annual operational energy
use, it produces a relatively small proportion of the overall carbon emissions since
it is generally provided by burning natural gas. In the UK, gas produces only 41%
of the carbon dioxide emissions per kilowatt hour that are produced by electricity.
Therefore the greatest scope for reducing carbon emissions from commercial buildings is in reducing the energy required for cooling, mechanical ventilation and artificial lighting.
Energy use is heavily dependent upon building design, including such factors as
layout, orientation, thermal capacity, glazing arrangements, solar shading, cooling
and ventilation. The choice of structural system (steel or concrete frame4) has been
found to have very little effect on operational energy requirements, so the following notes apply in general to all commercial buildings.
Structural steel gives the design team the flexibility it needs to exploit these
factors to the full.
One of the major ways of providing the necessary comfort levels is to make use
of fabric energy storage. This involves using the floor plate to absorb heat during
the period when the building is occupied during the day and then purging it at night
as shown in Fig. 8.3.
Initially there was a perception that thick concrete floors were needed for the
system to be effective. However, it has been shown that, with a 24-hour cycle of
heating and cooling, the necessary volume and disposition of concrete can be provided by composite construction.2
Perforated permeable ceilings can be used to permit the necessary contact
between the warm air and the floor. Enhancements can be achieved by ducting air
Steel Designers' Manual - 6th Edition (2003)
284
Sustainability and steel construction
Eli,
I
Night
Day
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Airflow
..'-_—-'_- Radiative heat transfer
Convective heat transfer
Fig. 8.3
Diurnal cycle for fabric energy storage
immediately above or below the slab. Examples are illustrated in the SCI publication Environmental Floor Systems.7
8.7 Summary
Steel in construction contributes in many ways to more sustainable solutions for the
built environment. The major points are that the basic material is 100% recyclable
without degradation, thereby contributing to conservation of resources. The building, in effect, becomes a warehouse of parts for the future.
Design and construction methods have been, and continue to be, refined to optimize the use of the material and the resources needed to process it. Steel structures
allow the creation of flexible space, and if they require alteration they lend themselves to adaptation.
At the end of their viable life on a particular site, appropriate design and detailing mean that components can be deconstructed and re-used rather than recycled.
A broader exposition of the issues raised can be found in the SCI Publication The
Role of Steel in Environmentally Responsible Buildings.6
References to Chapter 8
1. Eaton K.J. & Ogden R. (1995) Thermal and structural mass. The Architects
Journal, 202, No. 8, 24 Aug., 43.
2. Bennetts R. (2001) High watermark. The Architects Journal, 214, No. 19, 22 Nov.,
46–51.
Steel Designers' Manual - 6th Edition (2003)
References
285
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
3. Sansom M.R. (2001) Construction steel reuse and recycling rates approach 100%.
New Steel Construction, Vol. 9 No. 4, July/Aug., pp. 12–13.
4. Eaton K.J. & Amato A. (1998) A comparative environmental life cycle assessment
of modern office buildings. SCI P-182. The Steel Construction Institute, Ascot,
Berks.
5. The Building Regulations. (2002) Part L Conservation of Fuel and Power,
HMSO.
6. Gorgolewski M. (1999) The Role of Steel in Environmentally Responsible Buildings. SCI P-174. The Steel Construction Institute, Ascot, Berks.
7. Amato A., Ogden R. & Plank R. (1997) Environmental floor systems. SCI P-181.
The Steel Construction Institute, Ascot, Berks.
Steel Designers' Manual - 6th Edition (2003)
Chapter 9
Introduction to manual and
computer analysis
by RANGACHARI NARAYANAN
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
9.1 Introduction
The analysis of structures consists essentially of mathematical modelling of the
response of a structure to the applied loading. Such models are based on idealizations of the structural behaviour of the material and of the components. They are,
therefore, imperfect to a larger or smaller degree, depending upon the extent of
inaccuracy built into the assumptions in modelling. This is not to imply that the
calculations are meaningless, rather to emphasize the fact that the assessment of
structural responses is the best estimate that can be obtained in the light of the
assumptions implicit in the modelling of the system. Some of these assumptions
are necessary in the light of inadequate data; others are introduced to simplify the
calculation procedure to economic levels.
There are several idealizations introduced in the modelling process.
•
•
•
Firstly, the physical dimensions of the structural components are idealized.
For example, skeletal structures are represented by a series of line elements and
joints are assumed to be of negligible size. The imperfections in the member
straightness are ignored or at best idealized.
Material behaviour is simplified. For example, the stress–strain characteristic
is assumed to be linearly elastic, and then perfectly plastic. No account is taken
of the variation of yield stress along or across the member. The influence of
residual stresses due to thermal processes (such as hot rolling and flame cutting),
as well as that due to cold working and roller straightening, is ignored.
The implications of actions which are included in the analytical process itself
are frequently ignored. For example, the development of local plasticity at
connections or possible effects of change of geometry causing local instability
are rarely, if ever, accounted for in the analysis.
However, it must be recognized that the design loads employed in assessing structural response are themselves approximate. The analysis chosen should therefore
be adequate for the purpose and should be capable of providing the solutions at an
economical cost.
The fundamental concepts employed in mathematical modelling are discussed
next.
286
Steel Designers' Manual - 6th Edition (2003)
Element analysis
287
9.1.1 Equations of static equilibrium
From Newton’s law of motion, the conditions under which a body remains in static
equilibrium can be expressed as follows:
•
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
•
The sum of the components of all forces acting on the body, resolved along any
arbitrary direction, is equal to zero. This condition is completely satisfied if the
components of all forces resolved along the x, y, z directions individually add up
to zero. (This can be represented by SPx = 0, SPy = 0, SPz = 0, where Px, Py and
Pz represent forces resolved in the x, y, z directions.) These three equations
represent the condition of zero translation.
The sum of the moments of all forces resolved in any arbitrarily chosen plane
about any point in that plane is zero. This condition is completely satisfied when
all the moments resolved into xy, yz and zx planes all individually add up to zero.
(SMxy = 0, SMyz = 0 and SMzx = 0.) These three equations provide for zero
rotation about the three axes.
If a structure is planar and is subjected to a system of coplanar forces, the
conditions of equilibrium can be simplified to three equations as detailed below:
•
•
The components of all forces resolved along the x and y directions will individually add up to zero (SPx = 0 and SPy = 0).
The sum of the moments of all the forces about any arbitrarily chosen point in
the plane is zero (i.e. SM = 0).
9.1.2 The principle of superposition
This principle is only applicable when the displacements are linear functions of
applied loads. For structures subjected to multiple loading, the total effect of several
loads can be computed as the sum of the individual effects calculated by applying
the loads separately. This principle is a very useful tool in computing the combined
effects of many load effects (e.g. moment, deflection, etc.). These can be calculated
separately for each load and then summed.
9.2 Element analysis
Any complex structure can be looked upon as being built up of simpler units or
components termed ‘members’ or ‘elements’. Broadly speaking, these can be
classified into three categories:
•
Skeletal structures consisting of members whose one dimension (say, length) is
much larger than the other two (viz. breadth and height). Such a line element is
Steel Designers' Manual - 6th Edition (2003)
288
•
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
•
Introduction to manual and computer analysis
variously termed as a bar, beam, column or tie. A variety of structures are
obtained by connecting such members together using rigid or hinged joints.
Should all the axes of the members be situated in one plane, the structures so
produced are termed plane structures. Where all members are not in one plane,
the structures are termed space structures.
Structures consisting of members whose two dimensions (viz. length and breadth)
are of the same order but much greater than the thickness fall into the second
category. Such structural elements are called plated structures. Such structural
elements are further classified as plates and shells depending upon whether they
are plane or curved. In practice these units are used in combination with beams
or bars. Slabs supported on beams, cellular structures, cylindrical or spherical
shells are all examples of plated structures.
The third category consists of structures composed of members having all the
three dimensions (viz. length, breadth and depth) of the same order. The
analysis of such structures is extremely complex, even when several simplifying
assumptions are made. Dams, massive raft foundations, thick hollow spheres,
caissons are all examples of three-dimensional structures.
For the most part the structural engineer is concerned with skeletal structures.
Increasing sophistication in available techniques of analysis has enabled the economic design of plated structures in recent years. Three-dimensional analysis of
structures is only rarely carried out. Under incremental loading, the initial deformation or displacement response of a steel member is elastic. Once the stresses
caused by the application of load exceed the yield point, the cross section gradually
yields. The gradual spread of plasticity results initially in an elasto-plastic response
and then in plastic response, before ultimate collapse occurs.
9.3 Line elements
The deformation response of a line element is dependent on a number of crosssectional properties such as area, A, second moment of area (Ixx = Úy2 dA;
Iyy = Úx2 dA) and the product moment of area (Ixy = ÚxydA). The two axes xx and yy
are orthogonal. For doubly symmetric sections, the axes of symmetry are those for
which Úxy dA = 0. These are known as principal axes. For a plane area, the principal
axes may be defined as a pair of rectangular axes in its plane and passing through
its centroid, such that the product moment of area Úxy dA = 0, the co-ordinates
referring to the principal axes. If the plane area has an axis of symmetry, it is
obviously a principal axis (by symmetry Úxy dA = 0). The other axis is at right angles
to it, through the centroid of the area.
Tables of properties of the section (including the centroid and shear centre of
the section) are available as published data (e.g. SCI Steelwork Design Guide,
Vol. 1).1
Steel Designers' Manual - 6th Edition (2003)
Line elements
289
U
x
V
y
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 9.1
Angle section (no axis of symmetry)
If the section has no axis of symmetry (e.g. an angle section) the principal axes
will have to be determined. Referring to Fig. 9.1, if uOu and vOv are the principal
axes, the angle a between the uu and xx axes is given by
tan 2a =
-2 I xy
I xx - I yy
I uu =
I xx + I yy I xx - I yy
+
cos 2a - I xy sin 2a
2
2
(9.1)
I vv =
I xx + I yy I xx - I yy
cos 2a + I xy sin 2a
2
2
(9.2)
The values of a, Iuu and Ivv are available in published steel design guides (e.g.
Reference 1).
9.3.1 Elastic analysis of line elements under axial loading
When a cross section is subjected to a compressive or tensile axial load, P, the resulting stress is given by the load/area of the section, i.e. P/A. Axial load is defined as
one acting at the centroid of the section. When loads are introduced into a section
in a uniform manner (e.g. through a heavy end-plate), this represents the state of
stress throughout the section. On the other hand, when a tensile load is introduced
via a bolted connection, there will be regions of the member where stress concentrations occur and plastic behaviour may be evident locally, even though the mean
stress across the section is well below yield.
If the force P is not applied at the centroid, the longitudinal direct stress distribution will no longer be uniform. If the force is offset by eccentricities of ex and ey
measured from the centroidal axes in the y and x directions, the equivalent set of
actions are (1) an axial force P, (2) a bending moment Mx = Pex in the yz plane and
(3) a bending moment My = Pey in the zx plane (see Fig. 9.2). The method of evaluating the stress distribution due to an applied moment is given in a later section.
Steel Designers' Manual - 6th Edition (2003)
290
Introduction to manual and computer analysis
y
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
ey
y
Fig. 9.2
Compressive force applied eccentrically with reference to the centroidal axis
The total stress at any section can be obtained as the algebraic sum of the stresses
due to P, Mx and My.
9.3.2 Elastic analysis of line elements in pure bending
For a section having at least one axis of symmetry and acted upon by a bending
moment in the plane of symmetry, the Bernoulli equation of bending may be used
as the basis to determine both stresses and deflections within the elastic range. The
assumptions which form the basis of the theory are:
•
•
•
The beam is subjected to a pure moment (i.e. shear is absent). (Generally the
deflections due to shear are small compared with those due to flexure; this is not
true of deep beams.)
Plane sections before bending remain plane after bending.
The material has a constant value of modulus of elasticity (E) and is linearly
elastic.
The following equation results (see Fig. 9.3).
M f E
= =
I
y R
(9.3)
where M is the applied moment; I is the second moment of area about the neutral
axis; f is the longitudinal direct stress at any point within the cross section; y is the
distance of the point from the neutral axis; E is the modulus of elasticity; R is the
radius of curvature of the beam at the neutral axis.
Steel Designers' Manual - 6th Edition (2003)
-*
Line elements
neutral axis
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
cross section
Fig. 9.3
291
stress diagram
Pure bending
From the above, the stress at any section can be obtained as
f=
My
I
For a given section (having a known value of I) the stress varies linearly from zero
at the neutral axis to a maximum at extreme fibres on either side of the neutral axis:
M ymax M
=
I
Z
I
where Z =
.
ymax
fmax =
(9.4)
The term Z is known as the elastic section modulus and is tabulated in section tables.1
The elastic moment capacity of a given section may be found directly as the product
of the elastic section modulus, Z, and the maximum allowable stress.
If the section is doubly symmetric, then the neutral axis is mid-way between the
two extreme fibres. Hence, the maximum tensile and compressive stresses will be
equal. For an unsymmetric section this will not be the case, as the value of y for the
two extreme fibres will be different.
For a monosymmetric section, such as the T-section shown in Fig. 9.4, subjected
to a moment acting in the plane of symmetry, the elastic neutral axis will be the
centroidal axis. The above equations are still valid. The values of ymax for the two
extreme fibres (one in compression and the other in tension) are different. For an
applied sagging (positive) moment shown in Fig. 9.4, the extreme fibre stress in the
flange will be compressive and that in the stalk will be tensile. The numerical values
of the maximum tensile and compressive stresses will differ. In the case sketched in
Fig. 9.4, the magnitude of the tensile stress will be greater, as ymax in tension is greater
than that in compression.
Caution has to be exercised in extending the pure bending theory to asymmetric
sections. There are two special cases where no twisting occurs:
Steel Designers' Manual - 6th Edition (2003)
292
Introduction to manual and computer analysis
y
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 9.4
•
•
Monosymmetric section subjected to bending
Bending about a principal axis in which no displacement perpendicular to the
plane of the applied moment results.
The plane of the applied moment passes through the shear centre of the cross
section.
When a cross section is subjected to an axial load and a moment such that no
twisting occurs, the stresses may be determined by resolving the moment into components Muu and Mvv about the principal axes uu and vv and combining the resulting longitudinal stresses with those resulting from axial loading:
fu ,v = ±
P Muu v Mvv u
±
±
A
I uu
I vv
(9.5)
For a section having two axes of symmetry (see Fig. 9.2) this simplifies to
f x ,y = ±
P M xx y M yy x
±
±
A
I xx
I yy
Pure bending does not cause the section to twist. When the shear force is applied
eccentrically in relation to the shear centre of the cross section, the section twists
and initially plane sections no longer remain plane. The response is complex and
consists of a twist and a deflection with components in and perpendicular to the
plane of the applied moment. This is not discussed in this chapter. A simplified
method of calculating the elastic response of cross sections subjected to twisting
moments is given in an SCI publication.2
9.3.3 Elastic analysis of line elements subject to shear
Pure bending discussed in the preceding section implies that the shear force applied
on the section is zero. Application of transverse loads on a line element will, in
general, cause a bending moment which varies along its length, and hence a shear
force which also varies along the length is generated.
Steel Designers' Manual - 6th Edition (2003)
Line elements
b
293
shear stress
distribution
rectangular
cross section
(a)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
T
I
qmax)
-I
shear stress
distribution
I—section
(b)
Fig. 9.5
Shear stress distribution: (a) in a rectangular cross section and (b) in an I-section
If the member remains elastic and is subjected to bending in a plane of
symmetry (such as the vertical plane in a doubly symmetric or monosymmetric
beam), then the shear stresses caused vary with the distance from the neutral
axis.
For a narrow rectangular cross section of breadth b and depth d, subjected to a
shear force V and bent in its strong direction (see Fig. 9.5(a)), the shear stress varies
parabolically from zero at the lower and upper surfaces to a maximum value, qmax,
at the neutral axis given by
qmax =
3V
2bd
i.e. 50% higher than the average value.
For an I-section (Fig. 9.5(b)), the shear distribution can be evaluated from
q=
V
IB
y = hmax
Ú
by dy
(9.6)
y= h
where B is the breadth of the section at which shear stress is evaluated. The
integration is performed over that part of the section remote from the neutral axis,
i.e. from y = h to y = hmax with a general variable width of b.
Clearly, for the I- (or T-) section, at the web/flange interface the value of the
integral will remain constant. As the section just inside the web becomes the section
just inside the flange, the value of the vertical shear abruptly changes as B changes
from web thickness to flange width.
Steel Designers' Manual - 6th Edition (2003)
294
Introduction to manual and computer analysis
/
yield point
strain hardening
----<
strain hardening commences
0
strain
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 9.6
Idealized stress–strain relationship for mild steel
9.3.4 Elements stressed beyond the elastic limit
The most important characteristic of structural steels (possessed by no other
material to the same degree), is their capacity to withstand considerable deformation without fracture. A large part of this deformation occurs during the process of
yielding, when the steel extends at a constant and uniform stress known as the yield
stress.
Figure 9.6 shows, in its idealized form, the stress–strain curve for structural steels
subjected to direct tension. The line 0A represents the elastic straining of the
material in accordance with Hooke’s law. From A to B, the material yields while the
stress remains constant and is equal to the yield stress, fy. The strain occurring in
the material during yielding remains after the load has been removed and is called
plastic strain. It is important to note that this plastic strain AB is at least ten times
as large as the elastic strain, ey, at yield point.
When subjected to compression, various grades of structural steel behave in a
similar manner and display the same property of yield. This characteristic is known
as ductility of steel.
9.3.5 Bending of beams beyond the elastic limit
For simplicity, the case of a beam symmetrical about both axes is considered first.
The fibres of the beam subjected to bending are stressed in tension or compression
according to their position relative to the neutral axis and are strained as shown in
Fig. 9.7.
While the beam remains entirely elastic, the stress in every fibre is proportional
to its strain and to its distance from the neutral axis. The stress, f, in the extreme
fibres cannot exceed the yield stress, fy.
When the beam is subjected to a moment slightly greater than that which first
produces yield in the extreme fibres, it does not fail. Instead, the outer fibres yield
Steel Designers' Manual - 6th Edition (2003)
Line elements
(c)
(b)
(a)
295
(d)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 9.7 Elastic distribution of stress and strain in a symmetric beam. (a) Rectangular
section, (b) I-section, (c) stress distribution for (a) or (b), (d) strain distribution for
(a) or (b)
plastic zones
/ (comp.) \
plastic zones
(tension)
(a)
(b)
H
Y
(c)
strain distribution
(d)
Fig. 9.8 Distribution of stress and strain beyond the elastic limit for a symmetric beam.
(a) Rectangular section, (b) I-section, (c) stress distribution for (a) or (b), (d) strain
distribution for (a) or (b)
at constant stress, fy, while the fibres nearer to the neutral axis sustain increased
elastic stresses. Figure 9.8 shows the stress distribution for beams subjected to
such moments. Such beams are said to be partially plastic and those portions of their
cross sections which have reached the yield stress are described as plastic zones.
The depths of the plastic zones depend upon the magnitude of the applied
moment. As the moment is increased, the plastic zones increase in depth, and it is
assumed that plastic yielding will continue to occur at yield stress, fy, resulting in
two stress blocks, one zone yielding in tension and one in compression. Figure 9.9
represents the stress distribution in beams stressed to this stage. The plastic zones
occupy the whole area of the sections, which are then described as being fully plastic.
When the cross section of a member is fully plastic under a bending moment, any
attempt to increase this moment will cause the member to act as if hinged at that
point. This point is then described as a plastic hinge.
Steel Designers' Manual - 6th Edition (2003)
296
Introduction to manual and computer analysis
___ __
dEl nerfis
PH
_
fy 4
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(a)
(b)
(c)
(d)
Fig. 9.9 Distribution of stress and strain in a fully plastic cross section. (a) Rectangular
section, (b) I-section, (c) stress distribution for (a) or (b), (d) strain distribution for
(a) or (b)
The bending moment producing a plastic hinge is called the fully plastic moment
and is denoted by Mp. As the total compressive force and the total tensile force on
the cross section must be equal, it follows that the plastic neutral axis is also the
equal area axis, i.e. half the area of section is plastic in tension and the other half is
plastic in compression. This is true for monosymmetric or unsymmetrical sections
as well.
Shape factor
As described previously there will be two stress blocks, one in tension, the other in
compression, each at yield stress. For equilibrium of the cross section, the areas in
compression and tension must be equal. For a rectangular section the plastic
moment can be calculated as
Mp = 2 b
dd
bd 2
fy =
fy
2 4
4
which is 1.5 times the elastic moment capacity.
It will be noted that, in developing this increased moment, there is large straining in the external fibres of the section together with large rotations and deflections.
The behaviour may be plotted as a moment–rotation curve. Curves for various
sections are shown in Fig. 9.10.
The ratio of the plastic modulus, S, to the elastic modulus, Z, is known as the
shape factor, , and it will govern the point in the moment–rotation curve when
non-linearity starts. For the ideal section in bending, i.e. two flange plates, this
will have a value of unity. The value increases for more material at the centre of
the section. For a universal beam, the value is about 1.15 increasing to 1.5 for a
rectangle.
Steel Designers' Manual - 6th Edition (2003)
Line elements
297
(v1 .00)
1.00
0.87
(i'1.15)
0.67
(u=1.50)
M= jyMp
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
curvature
Fig. 9.10
Moment–rotation curves
Plastic hinges and rigid plastic analysis
In deciding the manner in which a beam may fail it is desirable to understand the
concept of how plastic hinges form when the beam becomes fully plastic. The
number of hinges necessary for failure does not vary for a particular structure
subject to a given loading condition, although a part of a structure may fail
independently by the formation of a smaller number of hinges. The member or
structure behaves in the manner of a hinged mechanism, and, in doing so, adjacent
hinges rotate in opposite directions.
As the plastic deformations at collapse are considerably larger than elastic ones,
it is assumed that the line element remains rigid between supports and hinge
positions i.e. all plastic rotation occurs at the plastic hinges.
Considering a simply-supported beam subjected to a point load at mid-span
(Fig. 9.11), the maximum strain will take place at the centre of the span where a
plastic hinge will be formed at yield of full section. The remainder of the beam will
remain straight: thus the entire energy will be absorbed by the rotation of the plastic
hinge.
= Mp(2q)
ÊL ˆ
Work done by the displacement of the load = W
q
Ë2 ¯
Work done at the plastic hinge
At collapse, these two must be equal:
WL
q
2
W = 4 M p L or
2 M pq =
M p = WL 4
The moment at collapse of an encastré beam with a uniformly distributed load
(w = W/L) is worked out in a manner similar to the above from Fig. 9.12.
Steel Designers' Manual - 6th Edition (2003)
Introduction to manual and computer analysis
298
w
plastic zone
yield zone
I
I__
L
I
lii
1!
I
stiff length
I
I
B
M
B
moment—rotation curve
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
bending moment diagram
Fig. 9.11
Centrally-loaded simply-supported beam
w/unit length
MA
I
I
I
(AI
I
C
L
I
I
I
I
/
)M8
loading
-i
Mp
B
collapse
M.
Fig. 9.12
Encastré beam with a uniformly distributed load
= Mp(q + 2q + q) = 4Mpq
WLL
WL
Work done by the displacement of the load (W ) =
q=
q
L 2 2
4
Work done at the three plastic hinges
Equating the two,
WL
q = 4 M pq
4
W = 16 M p L
or M p = WL 16
The moments at collapse for other conditions of loading can be worked out by a
similar procedure.
Steel Designers' Manual - 6th Edition (2003)
Line elements
299
9.3.6 Load factor and theorems of plastic collapse
The load factor at rigid plastic collapse, lp, is defined as the lowest multiple of the
design loads which will cause the whole structure, or any part of it, to become a
mechanism.
In the limit-state approach, the designer seeks to ensure that at the appropriate
factored loads the structure will not fail. Thus the rigid plastic load factor, lp, must
not be less than unity, under factored loads.
The number of independent mechanisms, n, is related to the number of possible
plastic hinge locations, h, and the degree of redundancy, r, of the skeletal structure,
by the equation
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
n=h-r
The three theorems of plastic collapse are given below for reference:
(1) Lower bound or static theorem
A load factor, ls, computed on the basis of an arbitrarily assumed bending
moment diagram which is in equilibrium with the applied loads and where the
fully plastic moment of resistance is nowhere exceeded, will always be less than,
or at best equal to, the load factor at rigid plastic collapse, lp.
lp is the highest value of ls which can be found.
(2) Upper bound or kinematic theorem
A load factor, lk, computed on the basis of an arbitrarily assumed mechanism
will always be greater than, or at best equal to, the load factor at rigid plastic
collapse, lp.
lp is the lowest value of lk which can be found.
(3) Uniqueness theorem
If both the above criteria ((1) and (2)) are satisfied, then l = lp.
9.3.7 Effect of axial load and shear
If a member is subjected to the combined action of bending moment and axial force,
the plastic moment capacity will be reduced.
The presence of an axial load implies that the sum of the tension and compression forces in the section is not zero (see Fig. 9.13). This means that the neutral axis
moves away from the equal area axis, providing an additional area in tension or
compression depending on the type of axial load. The presence of shear forces will
also reduce the moment capacity. For the beam sketched in Fig. 9.13,
axial load resisted = 2a t fy
Defining n =
2at
axial force resisted
=
,
axial capacity of section
A
Steel Designers' Manual - 6th Edition (2003)
Introduction to manual and computer analysis
300
B
T
total area = A
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 9.13
a=
f
total stresses = bending + axial compression
The effect of combined bending and compression
nA
2t
For a given cross section, the plastic moment capacity, Mp, can be evaluated as
explained previously. The reduced moment capacity, M¢p, in the presence of the axial
load can be calculated as follows:
M p¢ = M p - t a 2 fy
= Mp - t
n2 A2
n2 A2 ˆ
Ê
f
=
S
Á
˜ fy
y
Ë
4t ¯
4t 2
(9.7)
where S is the plastic modulus of the section.
Section tables provide the moment capacity for available steel sections using the
approach given above.1 Similar expressions will be obtained for minor axis bending.
9.3.8 Plastic analysis of beams subjected to shear
Once the material in a beam has started to yield in a longitudinal direction, it is
unable to sustain applied shear. When a shear, V, and an applied moment, M, are
applied simultaneously to an I-section, a simplifying assumption is employed to
reduce the complexity of calculations; shear resistance is assumed to be provided
by the web, hence the shear stress in the web is obtained as a constant value of V
divided by the web area (see Fig. 9.14). The longitudinal direct stress to cause yield,
f1, in the presence of this shear stress, q, is obtained by using the von Mises yield
criterion:3
f y2 = f 12 + 3q2
The reduced plastic moment capacity is given by
Ê fy - f1 ˆ
Mr = Mp - Á
˜M
Ë fy ¯ pw
(9.8)
Steel Designers' Manual - 6th Edition (2003)
Line elements
f
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
jd
(a)
Fig. 9.14
301
shear stress
longitudinal stress
(b)
(c)
Combined bending and shear
where Mpw is the fully plastic moment of resistance of the web.
The addition of an axial load to the above condition can be dealt with by
shifting the neutral axis, as was done in Fig. 9.13. The web area required to carry the
axial load is now given by P/f1 and the depth of the web, da, corresponding to this
is given by
da =
P
f1 t w
A further reduction in moment due to the introduction of the axial load is given
by
t w da2
f1
4
Hence the reduced moment capacity of the section is given by
t w da2
Ê fy - f1 ˆ
M1 = M p - Á
f1
˜ M pw Ë fy ¯
4
where f1 =
(9.9)
( fy2 - 3q 2 ) .
9.3.9 Plastic analysis for more than one condition of loading
When more than one condition of loading is to be applied to a line element, it may
not always be obvious which is critical. It is necessary then to perform separate
calculations, one for each loading condition, the section being determined by the
solution requiring the largest plastic moment.
Steel Designers' Manual - 6th Edition (2003)
302
Introduction to manual and computer analysis
Unlike the elastic method of design in which moments produced by different
loading systems can be added together, the opposite is true for the plastic theory.
Plastic moments obtained by different loading systems cannot be combined, i.e.
the plastic moment calculated for a given set of loads is only valid for that loading
condition. This is because the principle of superposition becomes invalid when parts
of the structure have yielded.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
9.4 Plates
Most steel structures consist of members which can be idealized as line elements.
However, structural components having significant dimensions in two directions
(viz. plates) are also encountered frequently. In steel structures, plates occur as components of I-, H-, T- or channel sections as well as in structural hollow sections.
Sheets used to enclose lift shafts or walls or cladding in framed structures are also
examples of plates.
With plane sheets, the stiffness and strength in all directions is identical and the
plate is termed isotropic. This is no longer true when stiffeners or corrugations are
introduced in one direction. The stiffnesses of the plate in the x and y directions are
substantially different. Such a plate is termed orthotropic.
The x and y axes for the analysis of the plate are usually taken in the plane of the
plate, as shown in Fig. 9.15, while the z axis is perpendicular to that plane. An
element of the plate will be subjected to six stress components: three direct stresses
(sx, sy and sz) and three shear stresses (txy, tyz and tzx). There are six corresponding
strains: three direct strains (ex, ey and ez) and three shear strains (gxy, gyz and gzx).
These stresses and strains are related in the elastic region by the material properties Young’s modulus (E) and Poisson’s ratio (v).
When considering the response of the plate, the approach customarily employed
is termed plane stress idealization. As the thickness, t, of the plate is small compared
with its other two dimensions in the x and y directions, the stresses having components in the z direction are negligible (i.e. sz, tyz and txz are all zero). This implies that
cTy
Fig. 9.15
Stress components on an element
Steel Designers' Manual - 6th Edition (2003)
Analysis of skeletal structures
303
the out-of-plane displacement is not zero, and this condition is referred to as plane
stress idealization.
For an isotropic plate, the general equation relating the displacement, w,
perpendicular to the plane of the plate element is given by
∂4 w
∂4 w
∂4 w q
+
2
+
=
∂x 4
∂x 2 ∂y 2 ∂y4 D
(9.10)
where q is the normal applied load per unit area in the z direction which will, in
general, vary with x and y. The term D is the flexural rigidity of the plate, given by
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
D=
E t3
12(1 - 2 )
(9.11)
The main difficulty in using this approach lies in the choice of a suitable
displacement function, w, which satisfies the boundary conditions. For loading
conditions other than the simplest, an exact solution of this differential equation is
virtually impossible. Hence approximate methods (e.g. multiple Fourier series) are
utilized. Once a satisfactory displacement function, w, is obtained, the moments per
unit width of the plate may be derived from
∂2 w ˆ
Ê ∂2 w
M x = - DÁ 2 + 2 ˜
Ë ∂x
∂y ¯
∂2 w ˆ
Ê ∂2 w
M y = - DÁ 2 + 2 ˜
Ë ∂y
∂x ¯
M xy = - M yx
(9.12)
∂2 w
= D(1 - )
∂x∂y
For orthotropic plates, the stiffness in x and y directions is different and the
equations are suitably modified as given below:
Dx
∂4 w
∂4 w
∂4 w
+
+
=q
2
D
D
xy
y
∂x 4
∂x 2 ∂y 2
∂y 4
(9.13)
where Dx and Dy are the flexural rigidities in the two directions.
In view of the difficulty of using classical methods for the solution of plate
problems, finite element methods have been developed in recent years to provide
satisfactory answers.
9.5 Analysis of skeletal structures
The evaluation of the stress resultants in members of skeletal frames involves the
solution of a number of simultaneous equations. When a structure is in equilibrium,
every element or constituent part of it is also in equilibrium. This property is made
use of in developing the concept of the free body diagram for elements of a structure.
Steel Designers' Manual - 6th Edition (2003)
304
Introduction to manual and computer analysis
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(0)
(b)
Fig. 9.16
Free body diagram
Fig. 9.17
Force and moments in x, y and z directions
The portal frame sketched in Fig. 9.16 will now be considered for illustrating the
concept. Assuming that there is an imaginary cut at E on the beam BC, the part
ABE continues to be in equilibrium if the two forces and moment which existed at
section E of the uncut frame are applied externally.The internal forces which existed
at E are given by (1) an axial force F, (2) a shear force V and (3) a bending moment
M. These are known as stress resultants. The external forces on ABE, together with
the forces F, V and M, keep the part ABE in equilibrium; Fig. 9.16(b) is called the
free body diagram. On a rigid jointed plane frame there are three stress resultants
at each imaginary cut. The part ECD must also remain in equilibrium. This
consideration leads to a similar set of forces F, V and M shown in Fig. 9.16(c). It
will be noted that the forces acting on the cut face E are equal and opposite. If the
two free body diagrams are moved towards each other, it is obvious the internal
forces F, V and M cancel out and the structure is restored to its original state of
equilibrium. As previously stated, equilibrium implies SPx = 0; SPy = 0; SM = 0 for
a planar structure. These equations can be validly applied by considering the structure as a whole, or by considering the free body diagram of a part of a structure.
In a similar manner, it can be seen that a three-dimensional rigid-jointed
frame has six stress resultants across each section. These are the axial force, two
shears in two mutually perpendicular directions and three moments, as shown in
Fig. 9.17.
Steel Designers' Manual - 6th Edition (2003)
Analysis of skeletal structures
305
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
With pin-jointed frames, be they two- or three-dimensional, there is only one
stress resultant per member, viz. its axial load. When forces act on an elastic structure, it undergoes deformations, causing displacements at every point within the
structure.
The solution of forces in the frames is accomplished by relating the stress resultants to the displacements. The number of equations needed is governed by the
degrees of freedom, i.e. the number of possible component displacements. At one
end of the member of a pin-jointed plane frame, the member displacement has
translational components in the x and y directions only, and no rotational displacement. The number of degrees of freedom is two. By similar reasoning it will be
apparent that the number of degrees of freedom for a rigid-jointed plane frame
member is three. For a member of a three-dimensional pin-jointed frame it is also
three, and for a similar rigid-jointed frame it is six.
9.5.1 Stiffness and flexibility
Forces and displacements have a vital and interrelated role in the analysis of structures. Forces cause displacements and the occurrence of displacements implies the
existence of forces. The relationship between forces and displacements is defined in
one of two ways, viz. flexibility and stiffness.
Flexibility gives a measure of displacements associated with a given set of forces
acting on the structure. This concept will be illustrated by considering the example
of a spring loaded at one end by a static load P (see Fig. 9.18).
As the spring is linearly elastic, the extension, D, produced is directly proportional to the applied load, P. The deflection produced by a unit load (defined as the
flexibility of the spring) is obviously D/P. Figure 9.18(b) illustrates the deflection
response of a beam to an applied load P. Once again the flexibility of the beam is
D/P.
In the simple cases considered above, flexibility simply gives the load–displacement response at a point. A more generalized definition applicable to the displace-
P
flexibility of
___________________
spring = 4flexibility of beam = 4-
(0)
Fig. 9.18
Flexibility
(b)
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
306
Fig. 9.19
Introduction to manual and computer analysis
Flexibility coefficients for a loaded beam
ment response at a number of locations will now be obtained by considering the
beam sketched in Fig. 9.19.
Considering a unit load acting at point 1 (Fig. 9.19(b)), the corresponding deflections at points 1, 2 and 3 are denoted as f11, f21 and f31 (the first subscript denotes
the point at which the deflection is measured; the second subscript refers to the
point at which the unit load is applied). The terms f11, f21, f31 are called flexibility coefficients. Figure 9.19(c) and (d) give the corresponding flexibility coefficients for load
positions 2 and 3 respectively. By the principle of superposition, the total deflections
at points 1, 2 and 3 due to P1, P2 and P3 can be written as
D1 = P1 f11 + P2 f12 + P3 f13
D2 = P1 f21 + P2 f22 + P3 f23
D3 = P1 f31 + P2 f32 + P3 f33
Written in matrix form, this becomes
ÏD 1 ¸ È f11 f12 f13 ˘ ÏP1 ¸
Ô Ô Í
˙Ô Ô
ÌD 2 ˝ = Í f21 f22 f23 ˙ ÌP2 ˝
ÔD Ô Í f f f ˙ ÔP Ô
Ó 3 ˛ Î 31 32 33 ˚ Ó 3 ˛
(9.14)
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Analysis of skeletal structures
307
k
Fig. 9.20
Stiffness coefficients
or
{D} = [F] {P}
where {D} = displacement matrix
[F] = flexibility matrix relating displacements to forces
{P} = force matrix
Hence {P} = [F]-1 {D}
Stiffness is the inverse of flexibility and gives a measure of the forces corresponding to a given set of displacements. Considering the spring illustrated in
Fig. 9.18(a), it is noted that the deflection response is directly proportional to
the applied load, P. The force corresponding to unit displacement is obviously P/D.
Likewise in Fig. 9.18(b) the load to be applied on the beam to cause a unit displacement at a point below the load is P/D. In its simplest form, stiffness coefficient
refers to the load corresponding to a unit displacement at a given point and can be
seen to be the reciprocal of flexibility. The concept is explained further using
Fig. 9.20.
First the locations 2 and 3 are restrained from movement and a unit displacement
is given at 1. This implies a downward force k11 at 1, an upward force k21 at 2 and a
downward force k31 at 3. The forces at points 2 and 3 are necessary as otherwise
there will be displacements at the locations 2 and 3.
The forces k11, k21 and k31 are designated as stiffness coefficients. In a similar
manner, the stiffness coefficients corresponding to unit displacements at points 2
and 3 are obtained.
Steel Designers' Manual - 6th Edition (2003)
308
Introduction to manual and computer analysis
The stiffness coefficients and the corresponding forces are linked by the following equations
P1
P2
P3
= k11D1 + k12D2 + k13D3
= k21D1 + k22D2 + k23D3
= k31D1 + k32D2 + k33D3
ÏP1 ¸ Èk11k12 k13 ˘ ÏD 1 ¸
Ô Ô
Ô Ô
or ÌP2 ˝ = Ík21k22 k23 ˙ ÌD 2 ˝
Í
˙
ÔP Ô Ík k k ˙ ÔD Ô
Ó 3 ˛ Î 31 32 33 ˚ Ó 3 ˛
(9.15)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
or {P} = [K]{D}
where [K] is the stiffness matrix relating forces and displacements.
9.5.2 Introduction to statically indeterminate skeletal structures
A structure for which the external reactions and internal forces and moments
can be computed by using only the three equations of statics (SPx = 0, SPy = 0 and
SM = 0) is known as statically determinate. A structure for which the forces
and moments cannot be computed from the principles of statics alone is statically
indeterminate. Examples of statically determinate skeletal structures are shown in
Fig. 9.21.
In structures shown in Fig. 9.21(a), (b) and (c), the supporting forces and moments
are just sufficient in number to withstand the external loading. For example, if one
of the supports of (b) were to fail or if one of the members of (c) were to be
removed, the structure would collapse.
However, when the beam or frame is provided with additional supports (see Fig.
9.21(d), (e)) or if the pin-jointed truss has more members than are required to make
it ‘perfect’ (Fig. 9.21(b)), the structure becomes statically indeterminate.
The degree of indeterminacy (also termed the degree of redundancy) is obtained
by the number of member forces or reaction components (viz. moments or
forces) which should be ‘released’ to convert a statically indeterminate structure
to a determinate one. If n forces or moments are required to be so released,
the degree of indeterminacy is n. We need n independent equations (in addition
to three equations of statics for a planar structure) to solve for forces and moments
at all locations in the structure. The additional equations are usually written by
considering the deformations or displacements of the structure. This means that
the section properties (viz. area, second moment of area, etc.) have an important
effect in evaluating the forces and moments of an indeterminate structure. Also,
the settlement of a support or a slight lack of fit in a pin-jointed structure contributes materially to the internal forces and moments of an indeterminate
structure.
Steel Designers' Manual - 6th Edition (2003)
Analysis of skeletal structures
309
Jr
C
D
A
B
VA
(b)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(0)
-
B
C
0
-HD
JjA
Jr
(d)
(c)
Al
(e)
(f)
Fig 9.21 Statically determinate and indeterminate skeletal structures. (a), (b) and (c) are
determinate; (d), (e) and (f) are indeterminate
Steel Designers' Manual - 6th Edition (2003)
310
Introduction to manual and computer analysis
9.5.3 The area moment method
The simplest technique of analysing a beam which is indeterminate to a low degree
is by the area moment method. The method is based on two theorems (see Fig. 9.22):
•
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
•
Area Moment Theorem 1: The change in slope (in radians) between two points
of the deflection curve in a loaded beam is numerically equal to the area under
the M/EI diagram between these two points.
Area Moment Theorem 2: The vertical intercept on any chosen line between the
tangents drawn to the ends of any portion of a loaded beam, which was originally straight and horizontal, is numerically equal to the first moment of the area
under the M/EI diagram between the two ends taken about that vertical line.
q B - q A = area of
M
diagram between A and B
EI
B
M
dx
EI
A
(9.16)
=Ú
D = moment of the
M
diagram between A and B taken about the
EI
vertical line RS
B
=
Mx
dx
EI
A
(9.17)
Ú
I
I
I
I
I
I
I
I
I
I M/EI diagram R
deflection curve ______
eB:eA
Fig. 9.22
Area moment theorems
Steel Designers' Manual - 6th Edition (2003)
Analysis of skeletal structures
311
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(Caution: The vertical intercept is not the deflection of the beam from its original
position.)
The area moment method can be used for solving problems like encastré beams,
propped cantilevers, etc. The procedure is as follows:
(1) The redundant supports are removed, thereby releasing the redundant forces
and moments. The statically determinate M/EI diagram for externally applied
loads can then be drawn.
(2) The externally applied loads are removed, the redundant forces and moments
are introduced one at a time, and the M/EI diagrams corresponding to each of
these forces and moments are drawn.
(3) The slopes at supports and intercepts on a vertical axis passing through the supports are then calculated.
(4) A number of expressions are obtained. These are then equated to known values
of slopes or displacements at supports. The equations so obtained can then be
solved for the unknown redundant reactions. This enables the evaluation of the
forces and moments in the structure.
9.5.4 The slope–deflection method
The slope–deflection method can be used to analyse all types of statically indeterminate beams and rigid frames. In this method all joints are considered rigid and
the angles between members at the joints are considered not to change as the loads
are applied. When beams or frames are deformed, the rigid joints are considered to
rotate as a whole.
In the slope–deflection method, the rotations and translations of the joints are
the unknowns. All end moments are expressed in terms of end rotations and translations. In order to satisfy the conditions of equilibrium, the sum of end moments
acting on a rigid joint must total zero. Using this equation of equilibrium, the
unknown rotation of each joint is evaluated, from which the end moments are
computed.
For the span AB shown in Fig. 9.23, the object is to express the end moments MAB
arid MBA in terms of end rotations qA and qB and translation D.
With the applied loading on the member, fixed end moments MFAB and MFBA are
required to hold the tangents at the ends fixed in direction. (Counter clockwise end
moments and rotations are taken as positive.) The slope–deflection equations for
the case sketched in Fig. 9.23 are
2 EI Ê
3D ˆ
-2q A - q B +
Ë
L
L¯
2 EI Ê
3D ˆ
= M FBA +
-2q B - q A +
Ë
L
L¯
M AB = M FAB +
M BA
(9.18)
Steel Designers' Manual - 6th Edition (2003)
312
Introduction to manual and computer analysis
CA
I
I
M
MFAB
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 9.23
Slope–deflection method
A
3M
(a)
Fig. 9.24
(b)
0
(c)
Moment-distribution method
where MFAB, MFBA are the fixed end moments at A and B due to loading and
settlement of supports (counter clockwise positive); qA and qB are end rotations;
and D is the downward settlement of support B relative to support A.
9.5.5 The moment-distribution method
The moment-distribution method can be employed to analyse continuous beams or
rigid frames. Essentially it consists of solving the simultaneous equations in the slope
–deflection method by successive approximations. Since the solution is by successive iteration, it is not even necessary to determine the degree of redundancy.
Two facets of the method must be appreciated (see Fig. 9.24):
•
•
When a stiff joint in a structural system absorbs an applied moment with
rotational movement only (i.e. no translation), Fig. 9.24(a), the moment resisted
by the various members meeting at the joint is in proportion to their respective
stiffnesses (Fig. 9.24(b)).
When a member is fixed at one end and a moment, M, is applied at the other
freely supported end, the moment induced at the fixed end is half the applied
Steel Designers' Manual - 6th Edition (2003)
Analysis of skeletal structures
313
A
positive moment
(clockwise)
negative moment
(counter clockwise)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
convention: clockwise moments are positive
convention: downward settlement of B with
respect to A is positive
Fig. 9.25
Sign conventions used in moment-distribution method
moment and acts in the same direction as M. (This is frequently referred to as
the carry over.) (Fig. 9.24(c).)
Figure 9.25 shows the sign conventions employed in the moment-distribution
method and Fig. 9.26 illustrates the moment-distribution procedure.
The moment-distribution method consists of locking all joints first and then
releasing them one at a time. To begin with, all joints are locked, which implies that
the fixed end moments due to applied loading will be applied at each joint. By
releasing one joint at a time, the unbalanced moment at each joint is distributed to
the various members meeting at the joint. Half of these applied moments are then
carried over to the other end of each member. This creates a further imbalance at
each joint and the unbalanced moments are once again distributed to all members
meeting at each joint in proportion to their respective stiffnesses.
This procedure is repeated until the totals of all moments at each joint are sufficiently close to zero. At this stage the moment-distribution process is stopped and
the final moments are obtained by summing up all the numbers in the respective
columns.
9.5.6 Unit load method
Energy methods provide powerful tools for the analysis of structures. The unit load
method can be directly derived from the complementary energy theorem, which
Steel Designers' Manual - 6th Edition (2003)
Introduction to manual and computer analysis
314
uniform load 12 kN/m
the problem
I uniform
4L1/L1
step 1: calculate the
distribution factors
stiffness factor for
3E1/L2
(4E1 + 3[\ (J + 3E1\
\T7 Ti \L1
L2I
BCi —
L2
A
2
2
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
as propped cantilever
r4:31
12
+!=
12
—16
+16
2
2
12
+!=
12
—16
+16
—1 6
(simply supported)
step 5: release joint B
+3.43
and distribute
0
+2.28
step 3: release joint C
step 4: carry over to B
—5
+4.57
step 2: calculate the
fixed end moments
caution: pay attention to signs
step 6: carry over to A
(note: no carry over to
C is possible)
+20.57
—13.72
Fig. 9.26
—20.57
0
step 7: sum up the moments
after repeating
distribution and carry
avers if necessary
An example of moment-distribution procedure
states that for any elastic structure in equilibrium under loads P1, P2, . . . , the
corresponding displacements x1, x2, . . . are given by the partial derivatives of the
complementary energy, C, with respect to the loads P1, P2, etc. In other words,
∂C
= x1
∂ P1
∂C
= x2
∂ P2
(9.19)
For a linearly elastic system, the complementary energy is equal to the strain energy,
U, hence
x1 =
∂U
∂ P1
x2 =
∂U
∂ P2
(9.20)
Steel Designers' Manual - 6th Edition (2003)
Analysis of skeletal structures
jPi
B
r2
jFs
C
D
315
iagram
11
B
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
IA
El
qram
Fig. 9.27
Unit load method
The total strain energy of an elastic system is given by the sum of the strain
energies stored in each member due to bending, shear, torsion and axial
loading.
The use of this will be illustrated by considering a simply-supported beam of
length l subject to an external loading (see Fig. 9.27). Strain energy stored in the
beam is predominantly flexural and is given by
l
M 2 dx
Ú 2EI
0
x1 = deflection under Pl =
l
l
∂ Ê M 2 dx ˆ
M ∂M
=
dx
Á
˜
∂ P1 Ë Ú0 2EI ¯ Ú0 EI ∂ P1
(9.21)
∂M
is the bending moment due to a unit load and is denoted by m.
∂ P1
Hence the procedure of the unit load method can be outlined (see Fig. 9.27):
M
diagram due to the external loading is obtained.
EI
(2) The external loads are now removed and the moment diagram (m) due to a
unit load applied at the point of required deflection is drawn.
(3) These two diagrams should now be integrated; in other words, the ordinates of
the two diagrams are multiplied to obtain the deflection, given by
(1)
l
x=Ú
0
M
m dx
EI
Steel Designers' Manual - 6th Edition (2003)
Introduction to manual and computer analysis
316
The same principle can be employed to determine the displacement due to other
causes, viz. axial load or shear or torsion.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
9.6 Finite element method
The advent of high-speed electronic digital computers has given tremendous
impetus to numerical methods for solving engineering problems. Finite element
methods form one of the most versatile classes of such methods which rely strongly
on the matrix formulation of structural analysis. The application of finite elements
dates back to the mid-1950s with the pioneering work by Argyris,4 Clough and
others.
The finite element method was first applied to the solution of plane stress
problems and subsequently extended to the analysis of axisymmetric solids,
plate bending problems and shell problems. A useful listing of elements developed
in the past is documented in text books on finite element analysis.5
Stiffness matrices of finite elements are generally obtained from an assumed displacement pattern. Alternative formulations are equilibrium elements and hybrid
elements. A more recent development is the so-called strain based elements. The
formulation is based on the selection of simple independent functions for the linear
strains or change of curvature; the strain–displacement equations are integrated to
obtain expressions for the displacements.
The basic assumption in the finite element method of analysis is that the response
of a continuous body to a given set of applied forces is equivalent to that of a system
of discrete elements into which the body may be imagined to be subdivided. From
the energy point of view, the equivalence between the body and its finite element
model is therefore exact if the strain energy of the deformed body is equal to that
of its discrete model.
The energy due to straining of the element, U, written in two-dimensional form
is
U=
1
(e xs x + e ys y + g xyt xy ) d(vol)
2 ÚÚ
or in matrix form
U=
1
T
{e } {s } d(vol)
2 ÚÚ
(9.22)
T
in which {e} = {e x , e y , g xy }
{s }
and {e}
{d c }
= {s x , s y , t xy }
T
(9.23)
= [ f ]{d c }
= {u1 , u2 , . . . , un }
T
(9.24)
Steel Designers' Manual - 6th Edition (2003)
Finite element method
317
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
where {d} is the nodal displacement vector, [f ] is a function defining the strain distribution and e refers to a typical finite element. When the strain distribution within
the model is exactly the same as that prevailing within the body, then the energy
equation will be exactly satisfied.
The exact determination of the strain distribution function [f ] in Equation (9.24)
presents considerable difficulties, since this can only be done by a rigorous solution
of the equations of linear elasticity. It may not always be possible to obtain an exact
shape function for the solution: however, a suitable function which is adequate to
model strains can usually be selected. The derivation of simple membrane elements
for plate problems is presented in the following pages.
9.6.1 Finite element procedure
As mentioned above, the basic concept of the finite element method is the
idealization of the continuum as an assemblage of discrete structural elements. The
stiffness properties of each element are then evaluated and the stiffness properties
of the complete structure are obtained by superposition of the individual element
stiffnesses. This gives a system of linear equations in terms of nodal point loads and
displacements whose solution yields the unknown nodal point displacements.
The idealization governs the type of element which must be used in the solution.
In many cases only one type of element is used for a given problem, but sometimes
it is more convenient to adopt a ‘mixed’ subdivision in which more than one type
of element is used.
The elements are assumed to be interconnected at a discrete number of nodal
points or nodes. The nodal degrees of freedom normally refer to the displacement
functions and their first partial derivatives at a node but very often may include
other terms such as stresses, strains and second or even higher partial derivatives.
For example, the triangular and rectangular membrane elements of Fig. 9.28(a) and
(b) have 6 and 8 degrees of freedom respectively, representing the translations u
and v at the corner nodes in the x and y directions.
A displacement function in terms of the co-ordinate variables x, y and the nodal
displacement parameters (e.g. ui, vi, or di) are chosen to represent the displacement
variations within each element. By using the principle of virtual work or the principle of minimum total potential energy, a stiffness matrix relating the nodal forces
to the nodal displacements can be derived. Hence the choice of suitable displacement functions is the most important part of the whole procedure. A good
displacement function leads to an element of high accuracy with converging
characteristics; conversely, a wrongly chosen displacement function yields poor or
non-converging results.
A displacement function may conveniently be established from simple polynomials or interpolation functions. The displacement field in each element must be
expressed as a function of nodal point displacements only, and this must be done in
such a way as to maintain inter-element compatibility, since this condition is neces-
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
318
Fig. 9.28
Introduction to manual and computer analysis
Triangular and rectangular membrane elements
sary to establish a bound on the strain energy. Therefore, the displacement pattern
and the nodal point degrees of freedom must be selected properly for each problem
considered.
In general, it is not always necessary that the compatibility must be satisfied in
order to achieve convergence to the true solution. If complete compatibility is not
achieved, there exists an uncertainty as to the bound on the strain energy of the
system. Therefore, in order to justify the performance and the ability of these elements to converge to the true solution, a critical test which indicates the performance in the limit must be carried out: that is, a convergence test with decreasing
mesh size. If the performance is adequate, then convergence to the true solution is
achieved within a reasonable computational effort. and the requirement for the
complete compatibility can be relaxed.
Besides satisfying the compatibility requirement, the assumed displacement
functions should include the following properties:
(1) Rigid body modes
(2) Constant strain and curvature states
(3) Invariance of the element stiffnesses.
9.6.2 Idealization of the structure
The finite element idealization should represent the real structure as closely as possible with regard to geometrical shape, loading and boundary conditions. The geometrical form of the structure is the major factor to be considered when deciding
the shape of elements to be used. In two-dimensional analyses the most frequently
used elements are triangular or rectangular shapes. The triangular element has the
advantage of simplicity in use and the ability to fit into irregular boundaries. Figure
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Finite element method
Fig. 9.29
319
Circular hole in uniform stress field
-
—a
Fig. 9.30
Perforated tension strip (plane stress) – a quarter of a plate is analysed
9.29 shows an example using triangular elements and Fig. 9.30 shows a combination
of triangular and rectangular elements. These figures also demonstrate the need to
use relatively small elements in areas where high stress gradients occur. In many
such cases the geometrical shape of the structure is such that a fine mesh of
elements is required in order to match this shape.
9.6.3 Procedure for evaluating membrane element stiffness
The formulation of the stiffness matrix [Kc] of the membrane element is briefly
discussed below.
The energy due to straining of the element is given by Equation (9.22). The stress
components are shown in Fig. 9.31 and ex, ey, gxy are the corresponding strain
Steel Designers' Manual - 6th Edition (2003)
Introduction to manual and computer analysis
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
320
Fig. 9.31
Stress components on a plane element
components. For the state of plane stress, the stress components are related to the
strain components as given below:
Ïs x ¸
E
Ô Ô
Ìs y ˝ =
2
Ôt Ô 1 - n
Ó xy ˛
È1 n
Í
Ín 1
Í
Í0 0
Î
0˘ e
˙Ï x ¸
0 ˙ Ôe Ô
Ì y ˝
1-n˙ Ô Ô
g
2 ˙˚ Ó xy ˛
(9.25)
where E and n are Young’s modulus and Poisson’s ratio respectively.
\ {s } = [D] {e}
The matrix [D] is referred to as the elasticity or property matrix. The strain energy
of the element is given by Equation (9.22) as
U=
1
T
{e} [D] {e} d(vol)
2 ÚÚ
(9.26)
Denoting the generalized nodal displacement by the vector {d e},
{d e } = [C ] {A}
(9.27)
where {A} = [a1, a2, . . . , an]T, which is a vector of polynomial constants, and [C] is a
transformation matrix.
The strains are obtained through making appropriate differentiations of the
displacement function with respect to the relevant co-ordinate variable x or y.
Thus,
Ïe x
Ô
{e } = Ìe y
Ôg
Ó xy
¸
Ô
˝=
Ô˛
Ï ∂u ∂ x ¸
Ô
Ô
Ô ∂v ∂ y Ô
˝
Ì
Ô ∂u + ∂v Ô
Ô ∂ y ∂x Ô
˛
Ó
(9.28)
Steel Designers' Manual - 6th Edition (2003)
Finite element method
\ {e} = [B]{A} = [B][C ]
-1
{d e }
321
(9.29)
where [B] is the transformation matrix.
Using Equations (9.26) and (9.29), the following expression for U in terms of the
nodal point displacement vector {d e} is obtained:
U=
1 e T
-1 ,T
{d } [C ]
2
[Ú Ú [B] [D][B]d(vol)][C ]
T
-1
{d e }
(9.30)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Differentiation of U with respect to the nodal displacements yields the stiffness
matrix [Ke]:
[K e ] = t[C ]
-1 ,T
È
˘ -1
T
Í Ú [B] [D][B]dA˙[C ]
ÎA
˚
(9.31)
where t is the thickness of the element (assumed constant).
The calculation of the stiffness matrices is generally carried out in two stages. The
first stage is to calculate the terms inside the square brackets of Equation (9.31) i.e.
the integration part. The second stage is to multiply the resulting integrations by the
inverse of the transformation matrix [C] and its transpose.
Equation (9.31) can now be written as
[K e ] = t[C ]
-1 ,T
[Q][C ]
-1
(9.32)
T
where [Q] = Ú [ B] [ D][ B]dA
A
The simplest elements for plane stress analysis have nodal points at the corners
only and have two degrees of kinematic freedom at each nodal point, i.e. u and v.
This type of element proves simple to derive and has been widely used. The
simplest elements of this type are rectangular and triangular in shape.
A triangular element with nodal points at the corners is shown in Fig. 9.28(a). The
displacement function of this element has two degrees of freedom at each nodal
point and the displacements are assumed to vary linearly between nodal points. This
results in constant values of the three strain components over the entire element;
the displacement functions are
u = a1 + a2x + a3y
v = a4 + a5x + a6y
(9.33)
The rectangular element with sides a and b, shown in Fig. 9.28(b), is used with the
following displacement functions:
u = a1 + a2x + a3y + a4xy
v = a5 + a6x + a7y + a8xy
(9.34)
Steel Designers' Manual - 6th Edition (2003)
Introduction to manual and computer analysis
322
x
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 9.32
yt'
Moments acting on a plane element
9.6.4 Procedure for evaluating plate bending element stiffness
The energy due to straining (bending) of the element is
Ub =
1
( c x M x + c y My + 2 c xy M xy )dA
2 ÚA
(9.35)
or in matrix form
ÏM x ¸
1
Ô
Ô
U b = Ú ( c x , c y , 2 c xy )ÌM y ˝dA
2A
ÔM Ô
Ó xy ˛
=
1
T
[ c ] {M}dA
2 AÚ
where the moments Mx, My, Mxy are given in Fig. 9.32 and cx, cy, and cxy are the
corresponding curvatures, i.e.
cx =
∂2 w
∂x2
cy =
∂2 w
∂ y2
c xy =
∂2 w
∂ x∂ y
(9.36)
where w is the transverse displacement of the plate element.
The conventional relationship between curvatures and moment is
ÏM x ¸
Ïc x ¸
Ô
Ô
Ô
Ô
M
=
D
[
]
Ì y ˝
Ìc y ˝
ÔM Ô
Ô2 c Ô
Ó xy ˛
Ó xy ˛
or
{M} = [D ]{c}
–
For an isotropic plate, the rigidity matrix, [D], may be written as
(9.37)
Steel Designers' Manual - 6th Edition (2003)
Finite element method
ÈD1 nD1
[D ] = ÍÍnD1 D1
0
ÎÍ0
323
0 ˘
0 ˙
˙
Dxy ˙˚
and
Dx = Dy = D1 =
Dxy =
Et 3
12(1 - n 2 )
1
(1 - n)D1
2
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Finally the strain energy can be written as
Ub =
1
T
[ c ] [D ][ c ]dA
2 ÚA
(9.38)
—
Denoting the generalized nodal displacement by the vector {d e },
{d } = [C ]{A}
e
(9.39)
–
where {A} is a vector of polynomial constants. From Equations (9.36) and (9.39),
{c} = [B ]{A} = [B ][C ] {d e }
-1
(9.40)
Using Equations (9.38) and (9.40), the following equation for Ub in terms of nodal
—
point displacement parameters {d e } is obtained:
Ub =
1 e
d
2
{ }
T
[C ]
-1 ,T
È
˘ -1 e
T
Í Ú [B ] [D ][B ]dA˙[C ] d
ÎA
˚
{ }
(9.41)
Differentiation of U with respect to the nodal displacements yields the stiffness
matrix [ K e]:
[K e ] = [C ]
-1 ,T
È
˘ -1
T
Í Ú [B ] [D ][B ]dA˙[C ]
ÎA
˚
(9.42)
Equation (9.42) can now be written as
[K e ] = [C ]
-1 ,T
[Q ][C ]
-1
(9.43)
where
T
[Q ] = Ú [B ] [D ][B ]dA
(9.44)
A
As mentioned before, the accuracy is dependent on choosing a large number of
elements. Many refined elements giving greater accuracy are described in standard
books on finite element methods, which also provide details of assembling the
Steel Designers' Manual - 6th Edition (2003)
324
Introduction to manual and computer analysis
elements and analysing the structure.5–7 These methods are used for solving a wide
range of problems.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
References to Chapter 9
1. The Steel Construction Institute (SCI) (2001) Steelwork Design Guide to BS
5950: Part 1: 2000, Vol. 1: Section Properties Member Capacities (6th edn) SCI,
Ascot, Berks.
2. Nethercot D.A., Salter P.R. & Malik A.S. (1989) Design of Members Subject to
Combined Bending and Torsion. The Steel Construction Institute, Ascot, Berks.
3. Timoshenko S. (1976) Strength of Materials – Part 2, 3rd edn. Van Nostrand &
Co., New York.
4. Argyris J.H. (1960) Energy Theorems and Structural Analysis. Butterworths,
London.
5. Zienkiewicz O.C. & Cheung Y.K. (2000) The Finite Element Method in Structural
and Continuum Mechanics, 5th edn. Butterworth-Heinemann, Oxford.
6. Coates R.C., Coutie M.G. & Kong F.K. (1988) Structural Analysis, 3rd edn.
Chapman & Hall, London.
7. Nath B. (1974) Fundamentals of Finite Elements for Engineers. Athlone Press,
London.
Further reading for Chapter 9
Brown D.G. (1995) Modelling of Steel Structures for Computer Analysis. The Steel
Construction Institute, Ascot, Berks.
Steel Designers' Manual - 6th Edition (2003)
Chapter 10
Beam analysis
by JOHN RIGHINIOTIS
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
10.1 Simply-supported beams
The calculations required to obtain the shear forces (SF) and bending moments
(BM) in simply-supported beams form the basis of many other calculations required
for the analysis of built-in beams, continuous beams and other indeterminate
structures.
Appropriate formulae for simple beams and cantilevers under various types of
loads are presented in the Appendix Bending moment, shear and deflection tables
for cantilevers and simply-supported beams.
In the case of simple beams it is necessary to calculate the support reactions
before the bending moments can be evaluated; the procedure is reversed for builtin or continuous beams. The following rules relate to the SF and BM diagrams for
beams:
(1) the shear force at any section is the algebraic sum of normal forces acting to
one side of the section
(2) shear is considered positive when the shear force calculated as above is upwards
to the left of the section
(3) the BM at any section is the algebraic sum of the moments about that section
of all forces to one side of the section
(4) moments are considered positive when the middle of a beam sags with respect
to its ends or when tension occurs in the lower fibres of the beam
(5) for point loads only the SF diagram will consist of a series of horizontal and
vertical lines, while the BM diagram will consist of sloping straight lines, changes
of slope occurring only at the loads
(6) for uniformly distributed loads (UDL) the SF diagram will consist of sloping
straight lines, while the BM diagram will consist of second-degree parabolas
(7) the maximum BM occurs at the point of zero shear, where such exists, or at the
point where the shear force curve crosses the base line.
10.2 Propped cantilevers
Beams which are built-in at one end and simply-supported at the other are known
as propped cantilevers. Normally, the ends of the beams are on the same level, in
325
Steel Designers' Manual - 6th Edition (2003)
326
Beam analysis
which case bending moments and reactions may be derived in two ways: by employing the Theorem of Three Moments or by deflection formulae. Appropriate formulae for propped cantilevers under various types of loading are presented in the
Appendix Bending moment, shear and deflection tables for propped cantilevers.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
10.2.1 Solution by the Theorem of Three Moments
Consider the propped cantilever AB in Fig. 10.1.
The bending moment at B may be found by using the Theorem of Three Moments,
and assuming that AB is one span of a two-span continuous beam ABC which is
symmetrical in every way about B.
Then the loads on AB and BC will produce free BM diagrams whose areas are
A1 and A2 respectively, the centres of gravity (CG) of the areas being distances x1
and x2 from A and C respectively.
M AL1 + 2 M B (L1 + L2 ) + M CL2 = 6
Now
Ê A1 x1 A2 x 2 ˆ
+
Ë L1
L2 ¯
where MA, MB and MC are the numerical values of the hogging moments at the supports A, B and C.
and
L1 = L2 = L
MA = MC = 0
A1 = A2 = A
x1 = x2 = x
Hence
2 M B (2L) = 6 ¥ 2
and
MB =
But
Ê Ax ˆ
Ë L ¯
3 Ax
L2
x
.— 1
area under
;;IlE:IJIT1IIIIILI:EtI:I;lz1:J:tJ:II1
A
Fig. 10.1
I—./1
B
Bending moment diagram for propped cantilever
\
C
Steel Designers' Manual - 6th Edition (2003)
Fixed, built-in or encastré beams
327
Therefore the moment at the fixed end of a propped cantilever
= 3 Ax L2
where A = the area of the free BM diagram, AB being considered as a simplysupported beam
x = the distance from the prop to the CG of the free BM diagram
and L = the span.
The reactions at each support may be found by employing a modified form of the
formula used for beams built-in at both ends:
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
SFA = the simple support reaction at A = SFB = the simple support reaction at B = +
MB
L
MB
L
where A is the propped end and B is built-in.
10.2.2 Sinking of supports
When the supports for a loaded propped cantilever do not maintain the same relative levels as in the unloaded condition, the BM and SF may be obtained by using
the deflection method (Fig. 10.2). When the prop, B, sinks the load which it takes is
reduced, while the fixing moment at the other end is increased. Two special cases
arise: the first when the prop sinks so much that no load is taken by the prop, and
the second when the built-in end sinks so much that the fixing moment is reduced
to zero, i.e. the cantilever resembles a simple support beam. The two special cases
are shown in Fig. 10.2.
10.3 Fixed, built-in or encastré beams
When the ends of a beam are firmly held so that they cannot rotate under the action
of the superimposed loads, the beam is known as a fixed, built-in or encastré beam.
The BM diagram for such a beam is in two parts: the free or positive BM diagram,
which would have resulted had the ends been simply-supported, i.e. free to rotate,
and the fixing or negative BM diagram which results from the restraints imposed
upon the ends of the beam.
Normally, the supports for built-in beams are at the same level and the ends of
the beams are horizontal. This type will be considered first.
Steel Designers' Manual - 6th Edition (2003)
328
Beam analysis
(a)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(b)
1L
I
I
I
I
I
BL,
IA4
-±
(c)
4RB
pIIIUUWWI
(a) + (c)
Fig. 10.2
(a) + (b)
Bending moment diagram for propped cantilever with sinking support
10.3.1 Beams with supports at the same level
The two conditions for solution, derived by Mohr, are:
(1) the area of the fixing or negative BM diagram is equal to that of the free or
positive BM diagram
(2) the centres of gravity of the two diagrams lie in the same vertical line, i.e. are
equidistant from a given end of the beam.
Figure 10.3 shows a typical BM diagram for a built-in beam.
ACDB is the diagram of the free moment Ms and the trapezium AEFB is the
diagram of the fixing moment Mi, the portions shaded representing the final
diagram.
Let As = the area of the free BM diagram
and Ai = the area of the fixing moment diagram.
Steel Designers' Manual - 6th Edition (2003)
Fixed, built-in or encastré beams
329
load
I
I
B
B
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 10.3
Bending moment diagram for fixed-end beam
Then from condition (1) above, As = Ai, while from condition (2) their centres of
gravity lie in the same vertical line, say, distance x from the left-hand support A.
Now
and
AE = the fixing moment MA
BE = the fixing moment MB.
Therefore
MA + MB
¥ L = Ai
2
and
MA + MB =
2 Ai
L
(10.1)
Divide the trapezium AEFB by drawing the diagonal EB and take area moments
about the support A.
Then
But
Also
Ê MA ¥ L L ˆ Ê MB ¥ L 2L ˆ
¥
+
¥
Ë
2
3¯ Ë
2
3 ¯
2
L
(MA + 2 MB )
=
6
6 Ai x
MA + 2 MB =
L2
Ai x
=
MA + MB =
2 Ai
L
(10.2)
(10.3)
As = Ai
Subtracting Equation (10.1) from Equation (10.2) and substituting As for Ai gives
Similarly
MB =
6 As x 2 As
L
L2
MA =
4 As 6 As x
- 2
L
L
Steel Designers' Manual - 6th Edition (2003)
330
Beam analysis
B
I
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
o _J
Fig. 10.4
Bending moment diagram for point load on fixed-end beam
It will be seen, therefore, that the fixing moments for any built-in beam on level supports can be calculated provided that the area of the free BM diagram and the position of its centre of gravity are known.
For point loads, however, the principle of reciprocal moments provides the
simplest solution.
With reference to Fig. 10.4,
Wab b Wab 2
¥ =
L
L
L2
Wab a Wa 2 b
MB =
¥ =
L
L
L2
MA =
i.e. the fixing moments are in reciprocal proportion to the distances of the ends of
the beam from the point load.
In the case of several isolated loads, this principle is applied to each load in turn
and the results summed.
It should be noted that appropriate formulae for built-in beams are given in the
Appendix Bending moment, shear and deflection tables for built-in beams.
10.3.2 Beams with supports at different levels
The ends are assumed, as before, to be horizontal.
The bent form of the unloaded beam as shown in Fig. 10.5 is similar to the bent
form of two simple cantilevers, which can be achieved by cutting the beam at the
centre C, and placing downward and upward loads at the free ends of the cantilevers
such that the deflection at the end of each cantilever is d/2.
Steel Designers' Manual - 6th Edition (2003)
Fixed, built-in or encastré beams
331
-
C
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
E
Fig. 10.5
Bending moment diagram for fixed-end beam with supports at different levels
Therefore
d P (L 2)
=
2
3EI
or
P=
3
(being the standard deflection formula)
12EId
L3
This load would cause a BM at A or B equal to
P¥
L 12EId L 6EId
=
¥ =
2
2
L3
L2
The solution in any given case consists of adding to the ordinary diagram of BM,
the BM diagram A1DCEB1.
Shear forces in fixed beams
In the case of fixed beams it is necessary to evaluate the BM before the SF can be
determined. This is the converse of the procedure for the case of simply-supported
beams.
The SF at the ends of a beam is found in the following manner:
SFA = the simple support reaction at A = +
MA - MB
L
SFB = the simple support reaction at B = +
MB - MA
L
where MA and MB are the numerical values of the moments at the ends of the beam.
These formulae must be followed exactly with respect to the signs shown since if
MA is smaller than MB the signs will adjust themselves.
Steel Designers' Manual - 6th Edition (2003)
Beam analysis
332
360 kN
LA
I
load
20m
4.Om
6.Om
bending
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
160 kNm
93.3 kN
1
]1320 kNm
moment
Ishear
12667
Fig. 10.6
Bending moment and shear force diagrams for fixed-end beam
It will be seen that for symmetrical loads, where MA = MB, the reactions will be
the same as for simply-supported beams.
An example of bending moment and shear force diagrams for a built-in beam
carrying a point load is given in Fig. 10.6.
10.4 Continuous beams
The solution of this type of beam consists, in the first instance, of the evaluation of
the fixing or negative moments at the supports.
The most general method is the use of Clapeyron’s Theorem of Three Moments.
The theorem applies only to any two adjacent spans in a continuous beam and in
its simplest form deals with a beam which has all the supports at the same level, and
has a constant section throughout its length.
The proof of the theorem results in the following expression:
M A ¥ L1 + 2 M B (L1 + L2 ) + MC ¥ L2 = 6
Ê A1 ¥ x1 A2 ¥ x2 ˆ
+
Ë L1
L2 ¯
Steel Designers' Manual - 6th Edition (2003)
Continuous beams
333
where MA, MB and MC are the numerical values of the hogging moments at the supports A, B and C respectively, and the remaining terms are illustrated in Fig. 10.7.
In a continuous beam the conditions at the end supports are usually known, and
these conditions provide starting points for the solution.
The types of end conditions are three in number:
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(1) simply supported
(2) partially fixed, e.g. a cantilever
(3) completely fixed, i.e. the end of the beam is horizontal as in the case of a fixed
beam.
The SF at the end of any span is calculated after the support moments have been
evaluated, in the same manner as for a fixed beam, each span being treated
separately.
It is essential to note the difference between SF and reaction at any support, e.g.
with reference to Fig. 10.7 the SF at support B due to span AB added to the SF at
B due to span BC is equal to the total reaction at the support.
If the section of the beam is not constant over its whole length, but remains
constant for each span, the expression for the moments is rewritten as follows:
MA ¥
L1
L2
Ê L1 L2 ˆ
Ê A1 ¥ x1 A2 ¥ x2 ˆ
+ 2 MB
+
+ MC ¥
=6
+
Ë
¯
Ë L1 ¥ I 1 L2 ¥ I 2 ¯
I1
I1 I 2
I2
any load system
fA
L1
I
x1
Fig. 10.7
L2
area A1
area A2
Clapeyron’s Theorem of Three Moments
c_f
Steel Designers' Manual - 6th Edition (2003)
334
Beam analysis
in which I1 is the second moment of area for span L1 and I2 is the second moment
of area for span L2.
Example
A two-span continuous beam ABC, of constant cross section, is simply supported
at A and C and loaded as shown in Fig. 10.8.
Applying Clapeyron’s theorem,
L1 = 2.0 m
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
L2 = 3.0 m
10 ¥ 2 2 ¥ 2 10
¥
=
kN m 2
8
3
3
200 ¥ 1 ¥ 2 3
¥ = 200 kN m 2
A2 =
3
2
A1 =
x1 = 1.0 m
200 kN
5kN/m
2.Om
2.Om
aria
3.Om
tOrn_tt
2.5
22.17
Fig. 10.8
I
L._.___.__j
32.17
IF 15.22
Bending moment and shear force diagrams for two-span beam
Steel Designers' Manual - 6th Edition (2003)
Plastic failure of single members
x2 =
335
4.0
m
3
Therefore M A ¥ 2 + 2 M B (2 + 3) + M C ¥ 3 = 6
Ê 10 ¥ 1 200 ¥ 4 ˆ
+
Ë 3¥2
3¥3 ¯
Since A and C are simple supports
MA = MC = 0
6 Ê 10 800 ˆ 6
(90.56) = 54.33 kN m
+
=
10 Ë 6
9 ¯ 10
0 - 54.33
SFA = 5 +
= 5 - 27.17 = -22.17 kN
2
SFB for span AB = 5 + 27.17 = 32.17 kN
200 ¥ 2 0 - 54.33
SFC =
+
= 133.33 - 18.11 = 115.22 kN
3
3
200
SFB for span BC =
+ 18.11 = 66.67 + 18.11 = 84.78 kN
3
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Therefore M B =
Note that the negative reaction at A means that the end A will tend to lift off its
support and will have to be held down.
10.5 Plastic failure of single members
The concept of the plastic hinge, capable of undergoing large rotation once the
applied moment has reached the limiting value Mp, constitutes the basis of plastic
design. This concept may be illustrated by examining the development of the collapse mode of a fixed-end beam subjected to a uniformly distributed load of increasing intensity w (Fig. 10.9(a)). Such a member is statically indeterminate, having three
redundancies which however reduce to two unknowns if the axial thrust in the
member is assumed to be zero. It will be assumed that the two unknown quantities
are the fixing moments MA and MB. As the load increases, the beam initially behaves
in an elastic manner and the value of the redundant moments can be derived by
applying the three general conditions used in elastic structural analysis, namely
those of
(1) equilibrium (application of statics)
(2) moment–curvature (EI d2y/dx2 = M)
(3) compatibility condition (continuity, including geometric conditions at the
supports).
For the beam in Fig. 10.9(a), the first condition (equilibrium) is satisfied by
drawing the bending moment diagram (shaded in Fig. 10.9(b)) as a superposition of
Steel Designers' Manual - 6th Edition (2003)
Beam analysis
336
tlength
P=o
P=o -t-—
MAt
L
f
MB
H
(a)
MAf
a1
b1
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
C
(b)
(c)
Fig. 10.9
Plastic failure of a fixed-end beam
the simply-supported sagging parabolic moment diagram ACB (peak value of
wL2/8) on the uniform moment diagram Aa1b1B due to the end hogging moments
MA and MB. From the moment–curvature relation applied over the full length of
the beam, it is readily deduced, by any one of a number of standard methods, that
the rotations of the end sections at A (qA clockwise) and at B (qB anti-clockwise)
and the central deflection DC are given by
qA =
wL3
MA L MB L
24EI 3EI
6EI
(10.4)
qB =
wL3
MA L MB L
24EI 6EI
3EI
(10.5)
DC =
5 wL4 M A L2 M B L2
384 EI
16EI
16EI
(10.6)
The compatibility conditions that now have to be satisfied are the directional
restraint of the end sections of the beam, i.e. qA = qB = 0, giving the well-known
Steel Designers' Manual - 6th Edition (2003)
Plastic failure of single members
337
result MA = MB = wL2/12. Equilibrium considerations now lead to the derivation of
the central sagging moment, MC. It follows from Fig. 10.9(b) that
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
MC =
wL2 Ê M A M B ˆ
+
Ë 2
8
2 ¯
(10.7)
whence MC = wL2/24, i.e. the end moments are twice the central moment. Finally,
application of Equation (10.3) gives the central deflection as wL4/384EI.
Suppose the load intensity w is increased until the fibres yield at some point in
the beam. It is assumed that the cross section has an idealized moment–curvature
relationship, i.e. the shape factor is unity (see Section 9.3.5). Up to this stage the
ratio of the end to central moments remains at 2, as represented by the dashed
line in Fig. 10.9(c). Due to the symmetry of the structure and loading, plastic hinges
will develop simultaneously at the fixed ends, i.e. at the points of maximum moment.
At the moment when the hinges form at the ends, the fixing moments have become
equal to the Mp of the beam and the loading intensity w has reached a value of
12Mp /L2. A slight load increase then causes the plastic hinges to rotate while
sustaining this constant moment Mp. This means that thereafter the beam behaves
as a simply-supported beam with constant end moments of Mp. The structure is
now statically determinate, and therefore the two degrees of redundancy in the
original problem no longer exist. In other words, two plastic hinges have formed,
eliminating a corresponding number of redundancies. At the same time two compatibility requirements have been eliminated; the condition that end slopes are zero
is no longer correct because the ends of the member are now rotating as plastic
hinges.
The central deflection at this stage, derived from Equation (10.6), has the value
DC =
5 Ê 12 M p ˆ L4 M p L2 M p L2
Á
˜
384 Ë L2 ¯ EI
8EI
32EI
and, from Equation (10.7), the central sagging moment MC becomes [(wL2/8) - Mp].
Further increases in the load intensity cause a third plastic hinge to form at midspan: see the full line in Fig. 10.9(c). This means that the moment MC attains a value
of Mp and therefore at C
Mp =
wL2
- Mp
8
hence
M p = wL2 16
or
w = 16 M p L2
The ratio of the end and central moments is now unity; as a result of the formation
of the plastic hinges there has been a redistribution of moments.
Substituting w = 16Mp/L2 and MA = MB = Mp in Equation (10.6) gives a central
deflection of DC = MpL2/12EI. When the final hinge has formed the deflection
increases rapidly without any further increase in the load. The beam is said to have
failed as a hinged mechanism.
Steel Designers' Manual - 6th Edition (2003)
338
Beam analysis
Now consider the same fixed-end beam with an initial settlement of wL4/144EI
at end A (see Fig. 10.10(a)). The results derived from moment–curvature considerations [Equations (10.1)–(10.3)] can still be applied provided qA, qB and DC are taken
with reference to the chord AB between the ends of the member, the chord having
rotated through an angle wL3/144EI relative to its initial position A0B.Thus the compatibility conditions at the ends of the member are now
q A = -q B = wL3 144EI
whence, by substitution in Equations (10.4), (10.5) and (10.7),
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
M A = wL2 24
M B = wL2 8
MC = wL3 24
The largest elastic moment occurs at B and therefore the first hinge starts there
at a load intensity given by Mp = wL2/8, i.e. w = 8Mp/L2. Equation (10.6) shows that
the central deflection is MpL2/144EI.
ABI
it length
e—
wL3
(0)
_____
01
b1
M
96
(b)
at '
b
(c)
Fig. 10.10
Plastic failure of a fixed-end beam with initial settlement
4
P
Steel Designers' Manual - 6th Edition (2003)
Plastic failure of propped cantilevers
339
The number of redundancies has now been reduced by one (since MB = Mp) while
the compatibility condition represented by Equation (10.7) no longer applies. Substitution of qA = wL3/144EI and MB = Mp in Equations (10.4) and (10.7) gives
5wL2 M p
48
2
2
7wL M p
MC =
96
4
MA =
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Using these new values for MA and MC the bending-moment diagram (Fig.
10.10(b)) is obtained, and inspection of the diagram shows that the second plastic
hinge can be expected to occur at end A. Putting MA = Mp gives
w=
3Mp
48 14.4 M p
¥ 2 =
2
5L
L2
MC = 0.8 M p
while Equation (10.3) gives DC = MpL2/16EI. The beam is now statically determinate with MA = MB = Mp, and eventually the third hinge forms at C when w reaches
16Mp /L2 (see Fig. 10.10(c)) and DC = MpL2/12EI. The important point to note is that,
despite the difference in initial conditions, the failure pattern, the failure load and
the deflection at the point of failure (relative to the ends) are the same for the two
cases. The uniqueness of the plastic limit load, i.e. its independence of initial conditions of internal stress or settlement of supports, is a general feature of plastic analysis. Deflections at the point of collapse can however be affected, as indeed they are
in the second case just described, when considered relative to the original support
position A0B (Fig. 10.10(a)).
10.6 Plastic failure of propped cantilevers
The case of the propped cantilever under a uniformly distributed load, Fig. 10.11(a),
cannot be solved quite so simply, and both upper and lower bound methods
described in Chapter 9 have to be used. A possible equilibrium condition is shown
in Fig. 10.11(b) where the reactant line a1B has been arranged so that the coordinate at the left-hand support is equal to the co-ordinate of the resultant moment
diagram at mid-span. At the right-hand support the condition of zero resultant
moment has to be satisfied. If the equal moments at A and C are regarded as plastic
hinge (Mp) values, the mechanism condition is satisfied. Hence Mp is an upper bound
(unsafe) value and should be denoted by Mu. Considering the geometry of the
moment diagram at mid-span,
lwL2
Mu
= Cc 1 + c 1c =
+ Mu
8
2
(M u = M p )
where l is the load factor at rigid plastic collapse (see Section 9.3.6)
Steel Designers' Manual - 6th Edition (2003)
340
Beam analysis
(a)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
I..
(b)
Fig. 10.11
H
(c)
Plastic failure of a propped cantilever
whence
Mc =
lwL2
= 0.0833lwL2
12
By a closer inspection of the resultant moment diagram, it is readily shown that the
maximum resultant moment does not occur at the mid-span, but between C and D,
at 5L/12 from the propped support. The value of this moment is lwL2/11.52 (=
0.0868lwL2), which is in excess of Mu. Note that this value is based on the moment
at the fixed end being lwL2/12. By making the plastic moment of the beam equal
to this higher value, a bending moment diagram satisfying the equilibrium and
plastic moment condition is derived, i.e. a static solution M1 = 0.0868lwL2. The
required Mp lies between these limits, M1 and Mu, i.e.
0.0833lwL2 £ 0.0868lwL2
To obtain the exact value of Mp, the sagging hinge is positioned at an unknown distance x from the right-hand support (Fig. 10.11(c)). Considering the total ordinate
of the free-moment diagram at this hinge position, then
x
lwL
lw 2
Mp + Mp =
xx
L
2
2
whence
Mp =
lwL Ê L - x ˆ
x
2 Ë L + x¯
(10.8)
Steel Designers' Manual - 6th Edition (2003)
Further reading
341
As the mechanism condition is satisfied, then any value of x inserted into
Equation (10.8) will give an upper bound solution, but the safest design will be
achieved when Mp is a maximum. By differentiating with respect to x, the solution
x = (÷2 - 1)L = 0.414L is obtained; substituting for x in Equation (10.5) gives Mp =
lwL2/1l.66 (= 0.0858lwL2). Note that this exact value of Mp lies within the range
indicated by the upper and lower bounds previously calculated.This particular result
is useful and can be applied directly to continuous beam problems where an end
span carries a uniformly distributed load.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Further reading for Chapter 10
Horne M.R. & Morris L.J. (1981) Plastic Design of Low-Rise Frames. Constrado
Monograph, Collins, London.
Kleinlogel A. (1948) Mehrstielige Rahmen, 6th edn. Ungar, New York.
Neal B.G. (1977) The Plastic Methods of Structural Analysis, 3rd edn. Chapman &
Hall, London.
Full Page Ad, 3mm Bleed
Steel Designers' Manual - 6th Edition (2003)
Chapter 11
Plane frame analysis
by JOHN RIGHINIOTIS
11.1 Formulae for rigid frames
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
11.1.1 General
The formulae given in this section are based on Professor Kleinlogel’s Rahmenformeln and Mehrstielige Rahmen.1 The formulae are applicable to frames which
are symmetrical about a central vertical axis, and in which each member has constant second moment of area.
Formulae are given for the following types of frame:
Frame I
Frame II
Frame III
Frame IV
Hingeless rectangular portal frame.
Two-hinged rectangular portal frame.
Hingeless gable frame with vertical legs.
Two-hinged gable frame with vertical legs.
The loadings are so arranged that dead, snow and wind loads may be reproduced
on all the frames. For example, wind suction acting normal to the sloping rafters of
a building may be divided into horizontal and vertical components, for which appropriate formulae are given, although all the signs must be reversed because the loadings shown in the tables act inwards, not outwards as in the case of suction.
It should be noted that, with few exceptions, the loads between node or panel
points are uniformly distributed over the whole member. It is appreciated that it is
normal practice to impose loads on frames through purlins, siderails or beams. By
using the coefficients in Fig. 11.1, however, allowance can be made for many other
symmetrically placed loads on the cross-beams of frames I and II shown, where the
difference in effect is sufficient to warrant the corrections being made. The indeterminate BMs in the whole frame are calculated as though the loads were uniformly distributed over the beam being considered, and then all are adjusted by
multiplying by the appropriate coefficient in Fig. 11.1. It may be of interest to state
why these adjustments are made. In any statically indeterminate structure the indeterminate moments vary directly with the value of the following quantity:
area of the free BM diagram
EI
Where the loaded member is of constant cross section, EI may be ignored.
342
Steel Designers' Manual - 6th Edition (2003)
I
I
liii
1111111
loa
IIII
III3I
1.00
0.60
parabola
0.75
total load = W
33131133
8 equal loads
1.111
total load = W
I I
I 1133
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
7 equal loads
1.125
total load = W
6 equal loads
1.143
total load = W
II IIII
5 equal loads
1.167
total load = W
4 equal loads
1.20
total load = W
II
3 equal loads
1.25
total load = W
2 equal loads
3W
total load
1.333
1.50
W
1.25
total load = W
ebolo
total load = W
_oI
IIIIiIIlIIII
!IIII!IIIIIIj
a(3L — 2a)
L2
L
total load = W
Fig. 11.1
1.20
Conversion coefficients for symmetrical loads
(3L2 — b2)
2L2
Steel Designers' Manual - 6th Edition (2003)
344
Plane frame analysis
Consider, as an example, the case of an encastré beam of constant cross section
and of length L carrying a UDL of W. Then the area of the free BM diagram is
WL 2L WL2
¥
=
8
3
12
If, however, W is a central point load, the area of the free BM diagram is
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
WL L WL2
¥ =
4
2
8
The fixed end moments (FEM) due to the two types of loadings are WL/12 and
WL/8 respectively, thus demonstrating that the indeterminate moments vary with
the area of the free BM diagram and proving that the indeterminate moments are
in the proportion of 1 : 1.5.
No rules can be laid down for the effect on the reactions of a change in the mode
of application of the load, although sometimes they will vary with the indeterminate moments. Consider a simple rectangular portal with hinged feet. If a UDL
placed over the whole of the beam is replaced by a central point load of the same
magnitude, then the knee moments will increase by 50% with a corresponding
increase in the horizontal thrusts H, while the vertical reactions V will remain the
same.
Although the foregoing remarks relating to the indeterminate moments resulting
from symmetrical loads apply to all rectangular portals, the rule applies for asymmetrical loads imposed upon the cross-beam of a rectangular portal frame with
hinged feet. If a vertical UDL on the cross-beam is replaced by any vertical load of
the same magnitude, then the indeterminate moments vary with the areas of the
respective free BM diagram.
No doubt readers who use the tables frequently will learn short cuts, but it is not
inappropriate to mention some. For example, if a UDL of W over the whole of a
single-bay symmetrical frame is replaced by a UDL of the same magnitude of W
over either the left-hand or right-hand half of the frame, the horizontal thrust at the
feet is unaltered. If the frame has a pitched roof then the ridge moment will also be
unaltered.
The charts in the Appendix have been prepared to assist in the design of rectangular frames or frames with a roof pitch of 1 in 5.
11.1.2 Arrangement of formulae
Each set of formulae is treated as a separate section. The data required for each
frame, together with the constants to be used in the various formulae, are given on
the first page. This general information is followed by the detailed formulae for the
various loading conditions, each of which is illustrated by two diagrams placed sideby-side, the left-hand diagram giving a loading condition and the right-hand one
giving the appropriate BM and reaction diagram. It should be noted, however, that
Steel Designers' Manual - 6th Edition (2003)
Formulae for rigid frames
345
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
some BMs change their signs as the frames change their proportions. This will be
appreciated by examining the charts.
For simple frames, i.e. for single-storey frames, the formulae for reactions immediately follow the formulae for BMs for each load.
Considering the simple frames only, the type of formula depends on the degree
of indeterminacy and the shape of the frame. Auxiliary coefficients X are introduced
whenever the direct expressions become complicated or for other reasons of
expediency.
No hard and fast rules can be laid down for the notation and it must be noted
that each set of symbols and constants applies only to the particular frame under
consideration, although, of course, an attempt has been made to produce similarity
in the types of symbols.
11.1.3 Sign conventions
All computations must be carried out algebraically, hence every quantity must
be given its correct sign. The results will then be automatically correct in sign and
magnitude.
The direction of the load or applied moment shown in the left-hand diagram for
each load condition is considered to be positive. If the direction of the load or
moment is reversed, the signs of all the results obtained from the formulae as printed
must be reversed.
For simple frames, the moments causing tension on the inside faces of the frame
are considered to be positive. Upward vertical reactions and inward horizontal reactions are also positive.
For multi-storey or multi-bay frames the same general rules apply to moments
and vertical reactions.
11.1.4 Checking calculations for indeterminate frames
Calculations for indeterminate frames may be checked by using some other method
of analysis, but it is also possible to check any frame or portion of a frame, such as
that above the line AB in Fig. 11.2, by ensuring that the following rules are obeyed:
(1) the three fundamental statical equations, i.e. SH = 0, SV = 0 and SM = 0, have
been satisfied, and, in addition, either that
(2) the sum of the areas of the M/EI diagram above any line, such as AB, is zero if
A and B are fully fixed; or
(3) the sum of the moments, with respect to the base AB, of the areas of the M/EI
diagram above the line AB is zero if A and B are partially restrained (as shown
in Fig. 11.2) or are hinged.
Steel Designers' Manual - 6th Edition (2003)
346
Plane frame analysis
-— side
boding
A
B
I
I—
1
I
'1
I
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
I
/
-7
L
Fig. 11.2
+
-7
I
I
/
Checking an indeterminate frame
The underlying principles in rules (2) and (3) above are those used in the application of the ‘column analogy’ method of analysis.
As an example of rule (2), consider the frame in Fig. 11.3 where EI is constant.
Then the sum of the areas of the M/EI diagram, considering the legs first, is
2 È (+0.0736 - 0.0826) ¥ 6.0 ˘ 2 È (-0.0826 + 0.0893) ¥ 8.078 ˘
˙˚ + EI ÍÎ
˙˚
EI ÍÎ
2
2
-0.054 + 0.054
=
=0
EI
thus demonstrating that the moments calculated are correct.
Now consider the frame in Fig. 11.4 as an example for rule (3).
Then the sum of the moments of the areas of the M/EI diagram, working from A
round to D, is
ÏÈ 11.25 ¥ 4.8 2 ¥ 4.8 ˘ È 6 ¥ 4.8 ¥ 2 4.8 ˘
¥
+
¥
ÌÍ
2
3 ˙˚ ÍÎ
3
2 ˙˚
ÓÎ
ÈÊ 11.25 - 12.75 ˆ
˘ È -12.75 ¥ 4.8 2 ¥ 4.8 ˘¸
+Í
9.6 ¥ 4.8˙ + Í
¥
˝=0
¯
2
2
3 ˙˚˛
ÎË
˚ Î
1
EI
1 È 4.8 2 ¥ 2
(11.25 + 6 - 4.5 - 12.75)˘˙ = 0
EI ÍÎ 2 ¥ 3
˚
demonstrating again that the calculations are correct.
11.2 Portal frame analysis
The design process explained here assumes that the reader is familiar with the basic
concepts of limit state steel design, and the basic analysis and design of continuous
Steel Designers' Manual - 6th Edition (2003)
Portal frame analysis
347
P
Checking a single-bay portal – rule (2)
0)
3
z
In
z
C-)
4.8 m
+
Fig. 11.4
2
m
w
I-
C
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
—4
Fig. 11.3
Checking a rectangular portal – rule (3)
beams. These notes take the form of a worked example. A method of determining
preliminary member sizes has been included.
11.2.1 Methods of analysis
BS 5950: Part 12 allows two main methods of analysis of a structure:
(1) Linear elastic. The frame is analysed either by hand or by computer assuming
linear elastic behaviour. Once the forces, moments and shears have been
derived by elastic analysis the ultimate capacity of each section is checked using
the rules given in Section 4 of the Code.
(2) Simple plastic theory. The frame is analysed using the basic principles of simple
plastic theory. Once the forces, moments and shears have been derived by analysis the member capacities are checked. Those containing plastic hinges are
checked in accordance with Section 4.
Elastic analysis
Using elastic analysis it is to be expected that the structure will be heavier than that
designed by plastic methods, but less stability bracing will be needed. It may well
be that the final details will also be more simple.
Steel Designers' Manual - 6th Edition (2003)
348
Plane frame analysis
It will remain the engineer’s responsibility to ensure that stability is provided both
locally and in the overall condition. BS 5950: Part 12 provides no specific rules
regarding the stability of the frame as a whole; this means that the engineer must
ensure that the stability is checked using the general rules for all frames. He must
also check that the movement of the frame under all loading cases is not sufficient
to cause damage to adjacent construction, i.e. brick walls or cladding, the serviceability limit state of deflection.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Plastic analysis
The method of calculating the ultimate load of a portal frame is described in many
publications. The main essence of the method is to assume that plastic ‘hinges’ occur
at points in the frame where the value of M/Mp is at its highest value, the load being
considered as increasing proportionally until the failure or ultimate state is reached.
Because of the straining at the hinge points it is essential that the local buckling and
lateral distortion do not occur before failure. Failure is deemed to have taken place
when sufficient hinges have formed to create a mechanism.
The member capacities are calculated using the rules given in Section 4 of the
Code but with the additional restrictions applied to hinge positions. In addition,
positive requirements are put on checking frame stability for both single-bay and
multi-bay frames. Plastic designed frames are lighter than elastic designed frames,
providing deflection is not a governing point; however, additional bracing may well
be required.
11.2.2 Stability
With the use of lighter frames, various aspects of stability take a more prominent
part in the design procedures. As far as portal frames are concerned the following
areas are important:
(1) overall frame stability, in that the strength of the frame should not be affected
by changes in geometry during loading (PD effect)
(2) snap-through stability, in multi-bay frames (three or more), where the effects of
continuity can result in slender rafters
(3) plastic hinge stability, where the member must be prevented from moving out
of plane or rotating at plastic hinges
(4) rafter stability, ensuring that the rafter is stable in bending as an unrestrained
beam
(5) leg stability, where the leg below the plastic hinge must be stable
(6) haunch stability, where the tapered member is checked to ensure that the inner
(compression flange) is stable.
Steel Designers' Manual - 6th Edition (2003)
Portal frame analysis
349
11.2.3 Selecting suitable members for a trial design
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
The design of a portal frame structure is in reality a process of selecting suitable
members and then proving their ability to perform in a satisfactory manner. Inexperienced engineers can be given some guidance to estimate initial member sizes.
In order to speed the initial selection of members, three graphs have been produced
to enable simple pin-based frames to be sized quickly.
These graphs have been prepared making the following assumptions:
(1) plastic hinges are formed at the bottom of the haunch in the leg and near the
apex in the rafter, the exact position being determined by the frame geometry
(2) the depth of the rafter is approximately span/55 and the depth of the haunch
below the eaves intersection is 1.5 times rafter depth
(3) the haunch length is 10% of the span of the frame, a limit generally regarded
as providing a balance between economy and stability
(4) the moment in the rafter at the top of the haunch is 0.87Mp, i.e. it is assumed
that the haunch area remains elastic
(5) the calculations assume that the calculated values of Mp are provided exactly
by the sections and that there are no stability problems. Clearly these conditions will not be met, and it is the engineer’s responsibility to ensure that the
chosen sections are fully checked for all aspects of behaviour.
The graphs cover the range of span/eaves height between 2 and 5 and rise/span of
0 to 0.2 (where 0 is a flat roof). Interpolation is permissible but extrapolation is not.
The three graphs give:
span/eaves height
2.0 2.5 3.0 3.5 4.0 4.5 5.0
\\\\\
0.2
0.15
C
0
0.1
0.05
0
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0,45
horzontaI force at base
Fig. 11.5
Rise/span against horizontal force at base for various span/eaves heights
Steel Designers' Manual - 6th Edition (2003)
350
Plane frame analysis
Figure 11.5: the horizontal force at the feet of the frame as a proportion of the total
factored load wL, where w is the load/unit length of rafter and L is the
span of the frame
Figure 11.6: the value of the moment capacity required in the rafters as a proportion of the load times span wL2
Figure 11.7: the value of the moment capacity required in the legs as a proportion
of the load times span wL2.
span/eaves height
5.0 4.0 3.0 2.0
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
0.2
0,1
0.05
0
0.02
0.03
0.04
0.045
P4 required for rafter
Fig. 11.6
Rise/span against required Mp of rafter for various span/eaves heights
II
span/eaves height
5.0 4.5 4.0 3.5 3.0
0.2
0.15
2.5
2.0
\\
0.05
\
N
0
0,045
0.05
0.055
0.06
0.065
0.07
0.075
0.08
Mp required for leg
Fig. 11.7
Rise/span against required Mp of leg for various span/eaves heights
Steel Designers' Manual - 6th Edition (2003)
Portal frame analysis
351
The graphs are non-dimensional and may be used with any consistent set of units.
In the worked example kilonewtons and metres are used.
Method of use of the graphs
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(1)
(2)
(3)
(4)
(5)
(6)
(7)
Determine the ratio span/height to eaves.
Determine the ratio rise/span.
Calculate wL (total load) and wL2.
Look up the values from the graphs.
Horizontal force at root of frame = value from Fig. 11.5 ¥ wL.
Mp required in rafter = value from Fig. 11.6 ¥ wL2.
Mp required in leg = value from Fig. 11.7 ¥ wL2.
11.2.4 Worked example of plastic design (see Fig. 11.8)
Determination of member sizes
Although the engineer may use his experience or other methods to determine preliminary member sizes, the graphs in Figs 11.5, 11.6 and 11.7 will be used for this
example. In order to use these graphs, four parameters are required:
3.75 m
7.60 m
25.00 m
frame centres 5.25 m
loading
unfactored
load factor
design
imposed
dead
0.75 kN/m2
0.43 kN/m2
1.6
1.4
total
factored load/m (w) = 237/25 = 9.48 kN/m
Fig. 11.8
Portal frame design example
factored
load
1.2 kN/m2
0.6 kN/m2
total factored
load
158 kN
79 kN
237 kN
Steel Designers' Manual - 6th Edition (2003)
352
Plane frame analysis
span/height to eaves
= 25/7.6
rise/span
= 3.75/25
wL (total load on frame) = 9.48 ¥ 25
= 3.29
= 0.15
= 237 kN
wL2
= 5925 kN m
= 9.48 ¥ 25 2
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Horizontal thrust at feet of frame (Fig. 11.5)
= 0.21 ¥ 237
= 49.8 kN m
Moment capacity of rafter (Fig. 11.6)
= 0.0305 ¥ 5925 = 181 kN m
Moment capacity of leg (Fig. 11.7)
= 0.059 ¥ 5925
= 350 kN m
Assuming a design strengh of 275 N/mm 2 (S275 steel):
S x required for rafter = 181 ¥ 1000/275 = 658 cm 3
S x required for leg
= 350 ¥ 1000/275 = 1270 cm 3
Trial sections (NB these are first trials and may not be adequate):
Rafter
406 ¥ 140 ¥ 39 kg UB (S275 steel);
S x = 721 cm 3 , I x = 12 500 cm 4
Leg
457 ¥ 152 ¥ 60 kg UB (S275 steel);
S x = 1280 cm 3 , I x = 25 500 cm 4
It is suggested that the next stage is to check the overall stability of the frame. The
main reason for this is that the only way to correct insufficient stability is to change
the main member sizes. If any other checks are not satisfied, additional bracing can
frequently be used to rectify the situation, without altering the member sizes.
References to Chapter 11
1. Kleinlogel A. (1931) Mehrstielige Rahmen. Ungar. New York.
2. British Standards Institution (2000) Structural use of steelwork in building. Part
1: Code of practice for design in simple and continuous construction: hot rolled
sections. BS 5950, BSI, London.
Further reading for Chapter 11
Baker J.F. (1954) The Steel Skeleton, Vol. 1, 1st edn. Cambridge University Press.
Steel Designers' Manual - 6th Edition (2003)
Further reading
353
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Baker J.F., Horne M.R. & Heyman J. (1956) The Steel Skeleton, Vol. 2. Cambridge
University Press.
Horne M.R. & Morris L.S. (1981) Plastic Design of Low-Rise Frames. Constrado
Monograph, Collins, London.
Neal B.G. (1977) The Plastic Methods of Structural Analysis, 3rd edn. Chapman &
Hall, London.
Steel Designers' Manual - 6th Edition (2003)
Chapter 12
Applicable dynamics
by MICHAEL WILLFORD
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
12.1 Introduction
The dynamic performance of steel structures has traditionally only been an area of
interest for special classes of structure, particularly slender wind-sensitive structures
(masts, towers and stacks), structures supporting mechanical equipment, offshore
structures and earthquake-resistant structures. However, a wider interest in dynamic
behaviour has recently come about as a result of the widespread adoption
of long span composite floors in buildings, for which vibration criteria must be
satisfied.
Dynamic loads in structures may arise from a number of sources including the
following:
Forces generated inside a structure – Machinery
– Impacts
– Human activity (walking, dancing, etc.)
External forces
– Wind buffeting and other aerodynamic
effects
– Waves (offshore structures)
– Impacts from vehicles, etc.
Ground motions
– Earthquakes
– Ground-borne vibration due to railways,
roads, pile driving, etc.
The principal effects of concern are:
Strength
– The structure must be strong enough to resist the peak dynamic forces
that arise.
Fatigue
– Fatigue cracks can initiate and propagate when large numbers of
cycles of vibration inducing significant stress are experienced, leading
to reduction in strength and failure.
Perception – Human occupants of a building can perceive very low amplitudes
of vibration, and, depending on the circumstances, may find vibration objectionable. Certain items of precision equipment are also
extremely sensitive to vibration. Perception will generally be the most
onerous dynamic criterion in occupied buildings.
354
Steel Designers' Manual - 6th Edition (2003)
Fundamentals of dynamic behaviour
355
It is beyond the scope of this book to address all these issues, and references are
suggested for more detailed guidance. The following sections are intended to give
an overview of the fundamental features of dynamic behaviour and to introduce
some of the terminology employed and analysis procedures available. It must be
borne in mind that dynamic behaviour is influenced by a larger number of parameters than static behaviour, and that some of the parameters cannot be predicted
precisely at the design stage. It is often advisable to investigate the effects of varying
initial assumptions to ensure that the most critical situations that may occur have
been examined.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
12.2 Fundamentals of dynamic behaviour
The principal features of dynamic behaviour may be illustrated by the examination
of a very simple dynamic system, a concentrated mass M supported on a light
cantilever of flexural rigidity EI and length L as illustrated in Fig. 12.1. The mass is
subjected to forces P which vary with time t.
12.2.1 Dynamic equilibrium
One of the basic methods of dynamic analysis is the examination of dynamic equilibrium to formulate an equation of motion. Consider the dynamic equilibrium of
the mass illustrated in Fig. 12.1. If at some time t the mass is displaced upwards from
its static equilibrium position by y, and has velocity y· and acceleration ÿ (positive
upwards) the mass is in general subjected to the following forces:
_—(
Ky
L
Forces acting on mass
M at time
Note: displacement y measured
relative to static
equilibrium position.
Displacement, velocity and
acceleration measured
positive upwards.
Fig. 12.1
Example of a dynamic system
Steel Designers' Manual - 6th Edition (2003)
356
Applicable dynamics
External force P(t)
Stiffness force Ky
Damping force Cy·
(K = 3EI/L for uniform cantilever)
(a dissipative force assumed to act in the opposite direction
to the velocity).
The resultant of these forces will cause the mass to accelerate according to Newton’s
2nd law of motion. Hence the dynamic equilibrium equation may be written as
My˙˙ = P (t ) - Ky - Cy˙
or
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
My˙˙ + Cy˙ + Ky = P (t )
(12.1)
Equation (12.1) is the general equation of motion for a single degree of freedom
dynamic system (a system whose behaviour can be defined by a single quantity, in
this case the deflection of the mass, y). It is clearly most important that the quantities in this equation are defined in dynamically consistent units. Examples of consistent units are given in Table 12.1.
Table 12.1
Mass
kg
tonnes
Force
Displacement
Time
N
kN
m
m
s
s
Solutions to Equation (12.1) for different assumptions regarding the force P(t)
can give an insight into the principal features of dynamic behaviour.
12.2.2 Undamped free vibration
In this case it is assumed that the system has been set into motion in some way and
is then allowed to vibrate freely in the absence of external forces. It is also assumed
that there is no damping.
The corresponding equation of motion derived from Equation (12.1) is:
Mÿ + Ky = 0
(12.2)
By putting K/M = w 2n it is easily shown that a motion of the form y = Y cos wnt satisfies this equation. This is known as simple harmonic motion; the mass oscillates
about its static equilibrium position with amplitude Y as shown in Fig. 12.2. wn is
known as the circular frequency, measured in radians per second. There are 2p
radians in a complete cycle of vibration and so the vibration frequency, fn, is wn/2p
cycles per second (hertz). In the absence of damping or external forces the system
will vibrate in this manner indefinitely.
Steel Designers' Manual - 6th Edition (2003)
Fundamentals of dynamic behaviour
357
time t (seconds)
Y COS Wnt
L.Tn=
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 12.2
Simple harmonic motion
The amplitude of vibration Y is the peak displacement of the mass relative to its
static equilibrium position. Vibration amplitudes are sometimes referred to as root
mean square (rms) quantities, where yrms = (1/T )÷(Ú T0 y 2 dt ), and T is the total time
over which the vibration is considered. For continuous simple harmonic motion yrms
= Y/√2.
12.2.3 Damped free vibration
Energy is always dissipated to some extent during vibration of real structures. Inclusion of the damping force in the free vibration equation of motion leads to
Mÿ + Cy· + Ky = 0
(12.3)
The solution to this equation for a lightly damped system when the mass is initially
displaced by Y and then released is
y = Ye -xw n t cos (w d t - f )
(12.4)
The motion takes the form shown in Fig. 12.3(a), and it can be seen that the vibration amplitude decays exponentially with time. The rate of decay is governed by the
amount of damping present.
If the damping constant C is sufficiently large then oscillation will be prevented
and the motion will be as in Fig. 12.3(b). The minimum damping required to prevent
overshoot and oscillation is known as critical damping, and the damping constant
for critical damping is given by Co = 2√(KM), where K and M are the stiffness and
mass of the system.
Practical structures are lightly damped, and the damping present is often
expressed as a proportion of critical. In Equation (12.4)
x = critical damping ratio =
C
Co
The phase shift angle f is small when the damping is small.
Steel Designers' Manual - 6th Edition (2003)
358
Applicable dynamics
Y
time
(a)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 12.3
(b)
Damped free vibration: (a) lightly damped system, (b) critically damped system
The frequency of damped oscillation is
1
w d = w n (1 - x 2 ) 2 radians/s
where wn is the undamped natural frequency. With low damping (x << 1) the reduction of natural frequency (increase in natural period) resulting from damping is
negligible.
12.2.4 Response to harmonic loads
A load which varies sinusoidally with time at a constant frequency is known as a
harmonic load. This form of dynamic load is characteristic of machinery operating
at constant speed, and many other types of continuous vibration can be approximated to this form.
When such a force of amplitude P and frequency f Hz (or w = 2pf radians/s) is
applied to the simple structure described in previous sections the structure will be
caused to vibrate. After some time the motions of the structure will reach a steady
state: that is, vibration of a constant amplitude and frequency will be achieved. The
dynamic equation of motion is:
Mÿ + Cy· + Ky = P cos wt
(12.5)
and it can be shown that the steady-state response motion is described by:
y = Y cos (w t - f )
(12.6)
Note that the steady-state vibration occurs at the frequency of the harmonic force
exciting the motion, not at the natural frequency of the structure.
The displacement amplitude Y can be shown to be:
Steel Designers' Manual - 6th Edition (2003)
Fundamentals of dynamic behaviour
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Y=
P
K
1
2 2
ÏÔÈ
w ˆ
Ê w ˆ ˘ Ê
ÌÍ1 - Ë
˙ + Ë 2x
¯
w
w
n
n ¯
˚
ÓÔÎ
2
¸Ô
˝
˛Ô
359
(12.7)
1
2
P/K is the deflection of the structure under a static force P, and the multiplying term
can be regarded as a magnification factor. The variation of magnification factor with
frequency ratio and damping ratio is shown in Fig. 12.4.
When the dynamic force is applied at a frequency much lower than the natural
frequency of the system (w/wn << 1), the response is quasi-static; the response is
governed by the stiffness of the structure and the amplitude is close to the static
deflection P/K.
When the dynamic force is applied at a frequency much higher than the natural
frequency (w/wn >> 1), the response is governed by the mass (inertia) of the structure; the amplitude is less than the static deflection.
When the dynamic force is applied at a frequency close to the natural frequency,
the stiffness and inertia forces in the vibrating system are almost equal and opposite at any instant, and the external force is resisted by the damping force. This is
the condition known as resonance, when very large dynamic magnification factors
are possible. When damping is low the maximum steady-state magnification occurs
when w wn, and then
Y=
P 1
K 2x
(12.8)
magnification factor
0
10
— 0.05
(note at resonance in lightly damped
structures magnification = 1
2
— 0.10
— 0.20
frequency ratio
0
0
Fig. 12.4
1.0
2.0
Steady-state response to harmonic loads
3.0
W
Steel Designers' Manual - 6th Edition (2003)
360
Applicable dynamics
Since in many structural systems x is of the order of 0.01, magnification factors
of the order of 50 may result. The force in the structure is proportional to the displacement so the same magnification factor applies to structural forces.
Human perception of motion is usually related to acceleration levels rather than
displacement. The peak acceleration amplitude at steady state is given by
â = w 2Y
2
= (2p f ) Y
(12.9)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
In practice, there is usually advantage in avoiding the possibility of resonance
whenever possible by ensuring that structural frequencies are well away from the
frequencies of any known sources of substantial dynamic force.
12.2.5 Response to an impact
Another dynamic loading case of interest is the response of a structure to an impact,
say from an object falling on to the structure. A full discussion of impact loading is
given in Reference 1, but a simple approximate method is useful for many practical situations when the mass of the impacting object is small compared with the
mass of the structure, and the impact duration is short compared with the natural
period of the structure. In these cases the effect of the impact can be assessed as an
impulse I acting on the structure. The magnitude of I may be calculated as m Dv,
where m is the mass of the falling object and Dv its change in velocity at impact.
If there is no rebound Dv can be taken as the approach velocity. For the simple
system discussed in previous sections conservation of momentum at impact requires
the initial velocity of the structural mass to be I/M. A lightly damped system then
displays damped free vibration corresponding to an initial displacement amplitude
of approximately
Y=
I
I
=
w n M 2 p fn M
(12.10)
12.2.6 Response to base motion
The previous sections have illustrated the behaviour of a dynamic system with a
fixed base subject to applied forces. When the dynamic excitation takes the form of
base motion, as for example in an earthquake, the formulation of the equation
of motion and the solutions are slightly modified. Detailed treatment of this type of
excitation is beyond the scope of this chapter and References 1 and 2 are recommended for discussion of the solutions.
Although not correct in detail, the general form of response indicated by Fig. 12.4
for harmonic applied forces is still relevant. Resonance occurs for those components
of the base motion close in frequency to the natural frequency of the structure.
Steel Designers' Manual - 6th Edition (2003)
Distributed parameter systems
361
12.2.7 Response to general time-varying loads
Although some dynamic loads (e.g. impacts and harmonic loads) are simple and can
be dealt with analytically, many forces and ground motions that occur in practice
are complex.
In general, numerical analysis procedures are required for the evaluation of
responses to these effects, and these are often available in finite element analysis
programs.
The techniques employed include:
•
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
•
•
direct step-by-step integration of the equations of motion using small time
increments,
Fourier analysis of the forcing function followed by solution for Fourier components in the frequency domain,
for random forces, random vibration theory and spectral analysis.
The background and application of these techniques are discussed in References 1
and 3.
12.3 Distributed parameter systems
The dynamic system considered in section 12.2 was simple in that its entire mass
was concentrated at one point. This enabled Newton’s second law (and therefore
the equation of motion) to be written in terms of a single variable.
Although some practical structures can be represented adequately in this way,
in other cases it is not realistic to assume that the mass is concentrated at one point.
The classical treatment of distributed systems is illustrated below by analysis
of a uniform beam; similar techniques can be used for two-dimensional (plate)
structures. The analysis calculates the natural frequency of the beam and, by reference to section 12.2.2, this can be done by ignoring external forces and damping
forces.
12.3.1 Dynamic equilibrium
The beam under consideration is shown in Fig. 12.5. For a small element of the beam
of length dx the variation of shear force V along the beam results in a net force on
∂V
the element of
dx which must cause the mass m dx to accelerate.
∂x
Dynamic equilibrium therefore requires:
∂V
∂2 y
dx = m dx 2
∂x
∂t
(12.11)
Steel Designers' Manual - 6th Edition (2003)
Applicable dynamics
362
small element
Ox
g th d x
M
(-*k) U+
(a)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Ox
dx
.x
V
Fig. 12.5
dx
(o)
Dynamic equilibrium of vibrating beams. (a) Uniform beam mass/unit length = m,
(b) small element mass = m dx
For a flexural beam the deflection y is a function of the bending moment distribution, with:
∂2 y M
=
∂ x 2 EI
noting also that for moment equilibrium on the element:
V=-
∂M
∂x
∂V
∂ 2M
=∂x
∂x 2
Substituting these relationships into Equation (12.11) leads to the equation of
motion for a flexural beam:
∂4y m ∂2y
+
=0
∂ x 4 EI ∂t 2
(12.12)
It is sometimes convenient to represent an entire framed structure as a shear beam.
In this case the deflection y is a function of the shear force, with
∂y
V
=
∂ x GAs
The equation of motion for a shear beam is therefore:
∂2y
m ∂2y
=0
∂ x 2 GAs ∂t 2
(12.13)
12.3.2 Modes of vibration
The partial differential equations derived above can be solved for given sets of
boundary conditions. The solution in each case identifies a family of modes of vibra-
Steel Designers' Manual - 6th Edition (2003)
Distributed parameter systems
363
tion (eigenvectors) with corresponding natural frequencies (eigenvalues). The mode
shape describes the relative displacements of the various parts of the beam at any
instant in time.
A number of standard results are given in Tables 12.2 and 12.3, and more comprehensive lists are contained in References 4 and 5.
Table 12.2 Natural frequencies of uniform flexural beams
Length =
Flexural rigidity =
Mass/unit length =
k
EI
fn = n
Hz:
2p mL4
Mode
(m) (typical units)
(N m2)
(kg/m)
values of k n given below
Shape
kn
Simply-supported
1
9.87
Nodal points at x/L =
0 1
2
39.5
0 0.5
3
88.8
0 0.333
(
x = sin np
n
1
x
L
)
y1
n2p 2
0
1
0.667
1
1
n -1
1
...
n
n
22.4
0 1
61.7
0 0.5
Encastré
x
2
3
121
1
Cantilever
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
( )
L
El
m
2
3
3.52
4 —.-\
1
0 0.359
0.641
0
22.0
0 0.774
61.7
0 0.5
0.868
1
Steel Designers' Manual - 6th Edition (2003)
364
Applicable dynamics
Table 12.3 Natural frequencies for uniform shear cantilevers
Length =
Shear rigidity =
Mass/unit length =
k
GAs
fn = n
Hz:
2p mL2
( )
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Mode
Shape
L
GAs
m
(m) (typical units)
(N)
(kg/m)
values of k n given below
kn
Nodal points at x/L =
1
1.57
0
2
4.71
0 0.667
3
7.85
0 0.4
(
n
y = sin (2n - 1)
px
2L
)
(2n - 1)
p
2
0
0.8
2
4
6
etc.
(2n - 1) (2n - 1) (2n - 1)
12.3.3 Calculation of responses
For linear elastic behaviour, once the mode shapes and frequencies have been established, dynamic responses can be calculated treating each mode as a single degree
of freedom system such as the one described in section 12.2.2. In theory, distributed
systems have an infinite number of modes of vibration, but in practice only a few
modes, usually those of lowest frequency, will contribute significantly to the overall
response.
It is convenient to describe a mode shape by a displacement parameter f defined
(or normalized) such that the maximum value at any point is 1.0. If the mode is
excited by dynamic forces resulting in a modal response amplitude of Ym, then the
displacement amplitude at a point i on the structure is
yi = fiYm
where fi is the mode shape value at the point.
The magnitude of Ym can be calculated for simple dynamic loads treating the
mode as a single degree of freedom system having the following properties. These
are often referred to as modal or generalized properties:
modal mass
M* = SMif i2
modal stiffness K* = w 2nM*
modal force
P* = SPifi
sum over whole structure
sum for all loaded points
Steel Designers' Manual - 6th Edition (2003)
Distributed parameter systems
365
Note that for mode n of a uniform simply-supported beam of span L and mass/unit
length m with f = sin(npx/L), the modal mass M* is:
L
2
M * = mÚ [sin (npx /L)] dx
0
which is 0.5 mL (half the total mass of the beam).
The application of these concepts to the problems of floor vibration and windinduced vibration is described in References 6 and 7.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
12.3.4 Approximate methods to determine natural frequency
Approximate methods are useful for estimating the natural frequencies of structures not conforming with one of the special cases for which standard solutions exist,
and for checking the predictions of computer analyses when these are used.
One of the most useful approximate methods relates the natural frequency of a
system to its static deflection under gravity load, d. With reference to section 12.2.2
the natural frequency of a single lumped mass system is:
fn =
1 Ê Kˆ
2p Ë M ¯
This may be rewritten, replacing the mass term M by the corresponding weight Mg,
as:
fn =
g
2p
Ê K ˆ
Ë Mg ¯
Since Mg/K is the static deflection under gravity load (d ):
fn =
g
2p d
= 15.76/ d
when d is measured in mm
(12.14)
This formula is exact for any single lumped mass system.
For distributed parameter systems a similar correspondence is found, although
the numerical factor in Equation (12.14) varies from case to case, generally between
16 and 20. For practical purposes a value of 18 will give results of sufficient
accuracy.
When applying these formulae the following points should be noted.
(1) The static deflection should be calculated assuming a weight corresponding to
the loading for which the frequency is required. This is usually a dead load with
an allowance for expected imposed load.
(2) For horizontal modes of vibration (e.g. lateral vibration of an entire structure)
the gravity force must be applied laterally to obtain the appropriate lateral
deflection, as shown in Fig. 12.6(a).
Steel Designers' Manual - 6th Edition (2003)
366
Applicable dynamics
—
—
—
/
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(0)
ft t tt fit fit t I tilt t
(b)
UUUU
(c)
Fig. 12.6
Use of gravity deflections to estimate natural frequencies. In each case fn 18/
√d. (a) For a vertical structure the gravity load is applied horizontally. (b) Continuous structure gravity load applied in opposite directions on alternate spans.
(c) Continuous structure – symmetric mode has higher frequency.
(3) The mode shape required must be carefully considered in multi-span structures.
In the two-span beam of Fig. 12.6(b) the lowest frequency will correspond to
an asymmetrical mode as illustrated; the corresponding d must be obtained by
applying gravity in opposite directions on the two spans. The normal gravity
deflection will correspond to the symmetrical mode with a higher natural frequency, Fig. 12.6(c).
These concepts can be extended to estimating the natural frequencies of primary
beam – secondary beam systems. If the static deflection of the primary beam is dp
and the static deflection of a secondary beam is ds (relative to the primary) the combined natural frequency is approximately
Steel Designers' Manual - 6th Edition (2003)
Damping
fn =
367
18
(d p + d s )
from which it can be shown that
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
1
1
1
= 2+ 2
2
fn fp f s
(12.15)
where fp and fs are the natural frequencies of the primary and secondary beams
alone.
Care must always be taken in using these formulae so that a realistic mode shape
is implied. There will generally be continuity between adjacent spans at small amplitudes even in simply-supported designs, and in many situations a combined mode
where both the primaries and secondaries are vibrating together in a ‘simplysupported’ fashion is not possible.
12.4 Damping
Damping arises from the dissipation of energy during vibration. A number of
mechanisms contribute to the dissipation, including material damping, friction
at interfaces between components and radiation of energy from the structure’s
foundations.
Material damping in steel provides a very small amount of dissipation and in
most steel structures the majority of the damping arises from friction at bolted
connections and frictional interaction with non-structural items, particularly partitions and cladding. Damping is found to increase with increasing amplitude of
vibration.
The amount of damping that will occur in any particular structure cannot be
calculated or predicted with a high degree of precision, and design values for
damping are generally derived from dynamic measurements on structures of a corresponding type.
Damping can be measured by a number of methods, including:
•
•
•
rate of decay of free vibration following an impact (Fig. 12.3(a))
forced excitation by mechanical vibrator at varying frequency to establish the
shape of the steady-state resonance curve (Fig. 12.4)
spectral methods relying on analysis of response to ambient random vibration
such as wind loading.
All these methods can run into difficulty when several modes close in frequency
are present. One result of this is that on floor structures (where there are often
several closely spaced modes) the apparent damping seen in the initial rate of decay
after impact can be substantially higher than the true modal damping.
Steel Designers' Manual - 6th Edition (2003)
368
Applicable dynamics
Table 12.4 Typical modal damping values by structure type
Structure type
Structure damping (% critical)
Unclad welded steel structures
(e.g. steel stacks)
Unclad bolted steel structures
Composite footbridges
Floor (fitted out), composite and
non-composite
0.3%
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Clad buildings (lateral sway)
0.5%
1%
1.5%–3%
(may be higher when many
partitions on floor)
1%
Damping is usually expressed as a fraction or percentage of critical (x), but the
logarithmic decrement (d ) is also used. The relationship between the two expressions is x = d/2p.
Table 12.4 gives typical values of modal damping that are suggested for use in calculations when amplitudes are low (e.g. for occupant comfort). Somewhat higher
values are appropriate at large amplitudes where local yielding may develop, e.g. in
seismic analysis.
12.5 Finite element analysis
Many simple dynamic problems can be solved quickly and adequately by the
methods outlined in previous sections. However, there are situations where more
detailed numerical analysis may be required and finite element analysis is a versatile technique widely available for this purpose. Numerical analysis is often necessary for problems such as:
(1) determination of natural frequencies of complex structures
(2) calculation of responses due to general time-varying loads or ground motions
(3) non-linear dynamic analysis to determine seismic performance.
12.5.1 Basis of the method
As explained in Chapter 9 the finite element method describes the state of a structure by means of deflections at a finite number of node points. Nodes are connected
by elements which represent the stiffness of the structural components.
In static problems the equilibrium of every degree of freedom at the nodes of the
idealization is described by the stiffness equation:
F = KY
where F is the vector of applied forces, Y is the vector of displacements for every
degree of freedom, and K is the stiffness matrix. Solution of unknown displacements
for a known force vector involves inversion of the stiffness matrix.
Steel Designers' Manual - 6th Edition (2003)
Finite element analysis
369
(0)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
-2jY3
_-0
1%
mosses at nodes
eam elements
connecting nodes
(b)
_____
/
(c)
Fig. 12.7
Finite element idealization. (a) Uniform beam in free vibration, (b) finite element
representation, (c) fourth mode not accurately represented
The extension of the method to dynamic problems can be visualized in simple
terms by considering the dynamic equilibrium of a vibrating structure.
Figure 12.7(a) shows the instantaneous deflected shape of a vibrating uniform
cantilever, and Fig. 12.7(b) shows a finite element idealization of this condition. The
shape is described by the deflections of the nodes, Y, and a mass is associated with
each degree of freedom of the idealization.
At the instant considered when the deflection vector is Y the forces at the nodes
provided by the stiffness elements must be KY.
These forces may in part be resisting external instantaneous nodal forces P and
may in part be causing the mass associated with each node to accelerate. The equations of motions of all the nodes may therefore be written as
P - MŸ = KY
or
KY + MŸ = P
where M is the mass matrix, Ÿ is the acceleration vector, and P is the external force
vector.
The natural frequencies and mode shapes are obtained by solving the undamped
free vibration equations:
Steel Designers' Manual - 6th Edition (2003)
370
Applicable dynamics
KY + MŸ = 0
Assuming a solution of the form y = Y cos wnt, it follows that Ÿ = -w 2nY and hence
[ K - w n2 M ]Y = 0
This is a standard eigenvalue problem of matrix algebra for which various numerical solution techniques exist. The solution provides a set of mode shape vectors Y
with corresponding natural frequencies wn. The number of modes possible will be
equal to the number of degrees of freedom in the solution.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
12.5.2 Modelling techniques
Dynamic analysis is more complex than static analysis and care is required so that
results of appropriate accuracy are obtained at reasonable cost when using finite
element programs. It is often advisable to investigate simple idealizations initially
before embarking upon detailed models. As problems and programs vary it is possible to give only broad guidance; individual program manuals must be consulted
and experience with the program being used is invaluable. More detailed background is given in Reference 8.
The first stage in any dynamic analysis will invariably be to obtain the natural frequencies and mode shapes of the structure. As can be seen from Fig. 12.7(c) a given
finite element model will represent higher modes with decreasing accuracy. If it is
only necessary to obtain a first mode frequency accurately then a relatively coarse
model, such as that illustrated in Fig. 12.7(b), will be perfectly adequate. In order to
obtain an accurate estimate of the fourth mode a greater subdivision of the structure would be necessary, since the distribution of inertia load along the uniform
beam in this mode is not well represented by just four masses.
Probably the most widely used approach for eigenvalue problems is subspace
iteration.This is a robust solution method which maps problems with a large number
of degrees of freedom on to a ‘subspace’, with a much smaller number of degrees
of freedom, to reduce the problem size. The structure’s eigenvalues (frequencies)
are the same as those of the subspace, and the eigenvectors (modeshapes) of the
structure are calculated from the eigenvectors of the subspace. This method resolves
the smallest eigenvalues with the highest accuracy. Most eigensolvers will also
include Sturm sequence checks to ensure that the eigenvalues found are the ones
required, and that there are none missing from the sequence.
Two approaches exist for calculating the element mass matrix. The consistent
mass matrix is the most accurate way of representing the mass and inertia of the
element, but when the aim is to calculate the dynamic response of the structure the
simpler lumped mass approach can be more effective as it avoids single element
modes of vibration.
It is important to ensure that all the relevant mass is accounted for in a modal
dynamic analysis. In many cases the engineer starts from a model, built for static
Steel Designers' Manual - 6th Edition (2003)
Dynamic testing
371
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
analysis, where ‘mass’ is applied as loading. It is important to ensure that all the
mass present is correctly included as mass (not loading) in the dynamic analysis.
Clearly, dynamically consistent units must be used throughout, and these units
may not be the same as those used for a static analysis using the same model. A
hand check of the first mode frequency using an approximate or empirical method
is strongly advisable to ensure that the results are realistic. In addition, there is even
more need than with static analysis to view computer analysis results as approximate. It is very difficult to predict natural frequencies of real structures with a high
degree of precision unless the real boundary conditions and structural stiffness can
be defined with confidence. This is rarely the case and these are uncertainties that
finite element analysis cannot resolve.
12.6 Dynamic testing
Calculation of the dynamic properties and dynamic responses of structures still presents some difficulties, and testing and monitoring of structures has a significant role
in structural dynamics. Testing is the only way by which the damping of structures
can be obtained, by which analytical methods can be calibrated and many forms of
dynamic loading can be estimated. It is often an essential part of the assessment and
improvement of structures where dynamic response is found to be excessive in practice. Further details are contained in references 9 and 10.
References to Chapter 12
1. Clough R.W. & Penzien J. (1993) Dynamics of Structures, 2nd edn. McGrawHill.
2. Dowrick D.J. (1987) Earthquake Resistant Design for Engineers and Architects,
2nd edn. John Wiley & Sons.
3. Warburton G.B. (1976) The Dynamical Behaviour of Structures, 2nd edn.
Pergamon Press, Oxford.
4. Harris C.M. & Crede C.E. (1976) Shock and Vibration Handbook. McGrawHill.
5. Roark R.J. & Young W.C. (1989) Formulas for Stress and Strain, 6th edn.
McGraw-Hill.
6. Construction Industry Research & Information Association (CIRIA)/The Steel
Construction Institute (SCI) (1989) Design Guide on the Vibration of Floors.
SCI Publication 076, SCI, Ascot, Berks.
7. Bathe K.J. (1996) Finite Element Procedures. Prentice Hall, Englewood Cliffs,
NJ.
8. Bachmann H. (1995) Vibration Problems in Structures. Birkhauser Verlag AG.
9. Ewins D.J. & Inman D.J. (2001) Structural dynamics @2000: Current status and
future directions. Research Studies Press, Baldock.
Steel Designers' Manual - 6th Edition (2003)
372
Applicable dynamics
10. Ewins D.J. (2001) Modal Testing: Theory, practice and application, 2nd edn.
Research Studies Press, Baldock.
Further reading for Chapter 12
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Blevins R.D. (1995) Formulas for natural frequency and mode shape (corrected
edition). Krieger, Malubar, FL.
Chopra A.K. (2001) Dynamics of Structures – Theory and Applications to Earthquake Engineering, 2nd edn. Prentice Hall, Upper Saddle River, NJ.
National Building Code of Canada (1995) Commentary A – Serviceability Criteria
for Deflections and Vibrations.
Steel Designers' Manual - 6th Edition (2003)
Chapter 13
Local buckling and
cross-section classification
by DAVID NETHERCOT
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
13.1 Introduction
The efficient use of material within a steel member requires those structural properties which most influence its load-carrying capacity to be maximized. This, coupled
with the need to make connections between members, has led to the majority of
structural sections being thin-walled as illustrated in Fig. 13.1. Moreover, apart from
circular tubes, structural steel sections (such as universal beams and columns, coldformed purlins, built-up box columns and plate girders) normally comprise a series
of flat plate elements. Simple considerations of minimum material consumption
frequently suggest that some plate elements be made extremely thin but limits
must be imposed if certain potentially undesirable structural phenomena are to
be avoided. The most important of these in everyday steelwork design is local
buckling.
Figure 13.2 shows a short UC section after it has been tested as a column. Considerable distortion of the cross-section is evident with the flanges being deformed
out of their original flat shape. The web, on the other hand, appears to be comparatively undeformed. The buckling has therefore been confined to certain plate elements, has not resulted in any overall deformation of the member, and its centroidal
axis has not deflected. In the particular example of Fig. 13.2, local buckling did not
develop significantly until well after the column had sustained its ‘squash load’ equal
to the product of its cross-sectional area times its material strength. Local buckling
did not affect the load-carrying capacity because the proportions of the web and
flange plates are sufficiently compact. The fact that the local buckling appeared in
the flanges before the web is due to these elements being the more slender.
Terms such as compact and slender are used to describe the proportions of the
individual plate elements of structural sections based on their susceptibility to local
buckling. The most important governing property is the ratio of plate width to plate
thickness, b, often referred to as the b/t ratio. Other factors that have some influence are material strength, the type of stress system to which the plate is subjected,
the support conditions provided, and whether the section is produced by hot-rolling
or welding.
Although the rigorous treatment of plate buckling is a mathematically complex
topic,1 it is possible to design safely and in most cases economically with no direct
consideration of the subject. For example, the properties of the majority of standard
hot-rolled sections have been selected to be such that local buckling effects are
373
Steel Designers' Manual - 6th Edition (2003)
374
Local buckling and cross-section classification
==
bb i
tJd
1b1
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
rolled beams
and columns
df9iJD
B
rhs
rolled
t
bIT
chs
tees
angles
Tb b
b=B—3t
d=D—3t
channels
T
T
b
b
T
df1r
In'
d[lHt
bT
___
b
I
HT T
Fig. 13.1
Structural cross-sections
I
1j
b
T
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Cross-sectional dimensions and moment–rotation behaviour
Fig. 13.2
375
Local buckling of column flange
unlikely to affect significantly their load-carrying capacity when used as beams or
columns. Greater care is, however, necessary when using fabricated sections for
which the proportions are under the direct control of the designer. Also, coldformed sections are often proportioned such that local buckling effects must be
accounted for.
13.2 Cross-sectional dimensions and moment–rotation behaviour
Figure 13.3 illustrates a rectangular box section used as a beam. The plate slenderness ratios for the flanges and webs are b/T and d/t, and elastic stress diagrams for
both components are also shown. If the beam is subject to equal and opposite end
moments M, Fig. 13.4 shows in a qualitative manner different forms of relationship
between M and the corresponding rotation q.
Assuming d/t to be such that local buckling of the webs does not occur, which of
the four different forms of response given in Fig. 13.4 applies depends on the compression flange slenderness b/T. The four cases are defined as:
(a) b/T £ b1, full plastic moment capacity Mp is attained and maintained for large
rotations and the member is suitable for plastic design – plastic cross-section
(Class 1).
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
376
Fig. 13.3
Local buckling and cross-section classification
Rectangular hollow section used as a beam
(b) b 1 < b/T < b 2, full plastic moment capacity Mp is attained but is only maintained
for small rotations and the member is suitable for elastic design using its full
capacity – compact cross-section (Class 2).
(c) b 2 < b/T £ b 3, full elastic moment capacity My (but not Mp) is attained and the
member is suitable for elastic design using this limited capacity – semi-compact
cross-section (Class 3).
(d) b 3 < b/T, local buckling limits moment capacity to less than My – slender crosssection (Class 4).
The relationship between moment capacity Mu and compression flange slenderness b/T indicating the various b limits is illustrated diagrammatically in Fig. 13.5.
In the figure the value of Mu for a semi-compact section is conservatively taken as
the moment corresponding to extreme fibre yield My for all values of b/T between
b 2 and b 3. This is more convenient for practical calculation than the more correct
representation shown in Fig. 13.4 in which a moment between My and Mp is
indicated. Since the classification of the section as plastic, compact, etc., is based on
considerations of the compression flange alone, the assumption concerning the web
slenderness d/t is that its classification is the same as or better than that of the flange.
For example, if the section is semi-compact, governed by the flange proportions,
then the web must be plastic, compact or semi-compact; it cannot be slender.
If the situation is reversed so that the webs are the controlling elements, then the
same four categories, based on the same definitions of moment–rotation behaviour,
are now determined by the value of web slenderness d/t. However, the governing
values of b 1, b 2 and b 3 change since the web stress distribution differs from the pure
compression in the top flange. Since the rectangular fully plastic condition, the triangular elastic condition and any intermediate condition contain less compression,
the values of b are larger. Thus section classification also depends upon the type of
Steel Designers' Manual - 6th Edition (2003)
Cross-sectional dimensions and moment–rotation behaviour
C
E
0
E
(b)
(a)
plastic
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
compact
rotation e
Fig. 13.4
Behaviour in bending of different classes of section
plastic
semi—
compact
/2
Ilange slenderness (b/I)
Fig. 13.5
slender
compact
Moment capacity as a function of flange slenderness
/3
377
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
378
Local buckling and cross-section classification
stress system to which the plate element under consideration is subjected. If, in addition to the moment M, an axial compression F is applied to the member, then for
elastic behaviour the pattern of stress in the web is of the form shown in Fig. 13.6(a).
The values of s1 and s2 are dependent on the ratio F/M with s2 approaching sy, if
F is large and M is small. In this case it may be expected that the appropriate b
limits will be somewhere between the values for pure compression and pure
bending, approaching the former if s2 ª sy, and the latter if s2 ª -s2. A qualitative
indication of this is given in Fig. 13.7, which shows Mu as a function of d /t for three
different s2/s1 ratios corresponding to pure compression, s2 = 0 and pure bending.
If the value of d /t is sufficiently small that the web may be classified as compact or
plastic, then the stress distribution will adopt the alternative plastic arrangement of
Fig. 13.6(b).
For a plate element in a member which is subject to pure compression the loadcarrying capacity is not affected by the degree of deformation since the scope for a
change in strain distribution as the member passes from a wholly elastic to a partially plastic state, as illustrated in Fig. 13.4 for pure bending, does not exist.
The plastic and compact classifications do not therefore have any meaning; the only
decision required is whether or not the member is slender, and specific values are
only required for b3.
stress due to compression F
stress due to moment M
(a)
stress due to compression F
stress due to moment M
a1 =
H
a2 =
(b)
Fig. 13.6
Stress distributions in webs of symmetrical sections subject to combined bending
and compression. (a) Semi-compact, elastic stress distribution. (b) Plastic or
compact, plastic stress distribution
Steel Designers' Manual - 6th Edition (2003)
Cross-sectional dimensions and moment–rotation behaviour
slender
compact
I compact______
plastic
______semi—
______
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
I
2
I
web slenderness (d/t)
13i
semislender
I
____ compact ____
____
plastic _______compact
/I
I
I2
/3 web slenderness (d/t)
MY111111i111N
a2
semi—
I
plastic
compact
I
compact
I
I
I
I?2
/33
slender
web slenderness (d/t)
Fig. 13.7
Moment capacity as a function of d/t for different web stress patterns
379
Steel Designers' Manual - 6th Edition (2003)
380
Local buckling and cross-section classification
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
In the introduction to this chapter several other factors which affect local
buckling are listed. These have a corresponding influence on b limits. As an example,
the flanges of an I-section receive support along one longitudinal edge only, with
the result that their buckling resistance is less than that of the flange of a box section,
and lower b values may be expected. Similarly the plate elements in members
fabricated by welding generally contain a more severe pattern of residual stress,
again leading to reduced b values.
One special case is the webs of beams and girders subject to shear. Although b
limits for the purpose of section classification are normally provided for designers,
the efficient design of plate girder webs may well require these to be exceeded.
Special procedures (see Chapter 17) are, however, normally provided for such
members.
13.3 Effect of moment–rotation behaviour on approach
to design and analysis
The types of member present in a structure must be compatible with the method
employed for its design. This is particularly important in the context of section
classification.
Taking the most restrictive case first, for a plastically designed structure, in which
plastic hinge action in the members is being relied upon as the means to obtain the
required load-carrying capacity, only plastic sections are admissible. Members which
contain any plate elements that do not meet the required b1 limit for the stress condition present are therefore unsuitable. This restriction could be relaxed for those
members in a plastically-designed structure not required to participate in plastic
hinge action: members other than those in which the plastic hinges corresponding
to the collapse mechanism form. However, such an approach could be considered
unsound on the basis of the effects of overstrength material, changes in the elastic
pattern of moments due to settlement or lack of fit, and so on. Something less than
a free choice of member types in structures designed plastically is required, and
BS 5950: Part 1 therefore generally restricts the method to structures in which only
plastic or compact sections are present.
When elastic design – in the sense that an elastically determined set of member
forces forms the basis for member selection – is being used, any of compact, semicompact or plastic sections may be used, provided member strengths are properly
determined. This point is discussed more fully in Chapters 14–18, which deal with
different types of member. As a simple illustration, however, for members subject
to pure bending the available moment capacity Mu must be taken as Mp or My. If
slender sections are being used the loss of effectiveness due to local buckling will
reduce not just their strength but also their stiffness. Moreover, reductions in stiffness are dependent upon load level, becoming greater as stresses increase sufficiently to cause local buckling effects to become more significant. Strictly speaking,
such changes should be included when determining member forces. However, any
attempt to do this would render design calculations prohibitively difficult and it is
Steel Designers' Manual - 6th Edition (2003)
Economic factors
381
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
therefore usual to make only a very approximate allowance for the effect. It is of
most importance for cold-formed sections.
In practice the designer, having decided upon the design approach (essentially
either elastic or plastic), should check section classification first using whatever
design aids are available. Since most hot-rolled sections are at least compact in
both S275 and S355 steel, this will normally be a relatively trivial task. When using
cold-formed sections, which will often be slender, sensible use of manufacturer’s
literature will often eliminate much of the actual calculation. Greater care is
required when using sections fabricated from plate, for which the freedom to select
dimensions and thus b/T and d/t ratios means that any class is possible.
13.4 Classification table
Part of a typical classification table, extracted from BS 5950: Part 1, is given in Table
13.1. Values of b 1, b 2, and b 3 for flanges, defined as plates supported along one longitudinal edge, and webs, defined as plates supported along both longitudinal edges,
under pure compression, pure bending and combined compression and bending are
listed. While the third case reduces to the second as the compression component
reduces to zero, it does not accord with the first case when the web is wholly subject
to uniform compression. The reason for this is that the neutral axis of a member
subject to bending and compression in which the web is wholly in compression must
lie in the flange, or at least at the web/flange junction, with the result that the tensile
strains in the flange provide some degree of stabilizing influence. A slightly higher
set of limits than those provided for a plate supported by other elements which are
themselves in compression, such as the compression flange of a box beam, is therefore appropriate.
13.5 Economic factors
When design is restricted to a choice of suitable standard hot-rolled sections, local
buckling is not normally a major consideration. For plastically designed structures
Table 13.1 Extract from table of section classification limits (BS 5950: Part 1)
Type of element
Class of section
Plastic (b 1)
Outstand element of compression flange
Internal element of compression flange
Web with neutral axis at mid-depth
Web, generally
b/T
b/T
d/t
d/t
£ 9e
£ 28e
£ 80e
£ 80e /(1 + r1)
Compact (b 2)
b/T
b/T
d/t
d/t
£
£
£
£
10e
32e
100e
100e /(1 + r1)
Semi-compact (b 3)
b/T
b/T
d/t
d/t
£
£
£
=
15e
40e
120e
120e /(1 + r1)
in which r1 = Fc /dtpyw lies between -1 and +1 and is a measure of the stress ratio within the web
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
382
Local buckling and cross-section classification
only plastic sections are suitable: thus the designer’s choice is slightly restricted,
although no UBs and only 4 UCs in S275 steel and 7 UBs and 9 UCs in S355 steel
are outside the limits of BS 5950: Part 1 when used in pure bending. Although considerably more sections are unsatisfactory if their webs are subject to high compression, the number of sections barred from use in plastically designed portal
frames is, in practice, extremely small. Similarly for elastic design no UB is other
than semi-compact or better, provided it is not required to carry high compression
in the web, while all UCs are at least semi-compact even when carrying their full
squash load.
The designer should check the class of any trial section at an early stage. This can
be done most efficiently using information of the type given in Reference 2. For
webs under combined compression and bending the first check should be for pure
compression as this is the more severe. Provided the section is satisfactory no additional checks are required; if it does not meet the required limit a decision on
whether it is likely to do so under the less severe combined load case must be made.
The economic use of cold-formed sections, including profiled sheeting of the type
used as decking and cladding, often requires that members are non-compact. Quite
often they contain plate elements that are slender, with the forming process being
exploited to provide carefully proportioned shapes. Since cold-formed sections are
proprietary products, manufacturers normally provide design literature in which
member capacities which allow for the presence of slender plate elements are listed.
If rigorous calculations are, however, required, then Parts 5 and 6 of BS 5950 contain
the necessary procedures.
When using fabricated sections the opportunity exists for the designer to optimize on the use of material. This leads to a choice between three courses of action:
(1) eliminate all considerations of local buckling by ensuring that the width-tothickness ratios of every plate element are sufficiently small;
(2) if employing higher width-to-thickness ratios, use stiffeners to reduce plate
proportions sufficiently so that the desired strength is achieved;
(3) determine member capacities allowing for reductions due to exceeding the
relevant compact or semi-compact limits.
Effectively only the first of these is available if plastic design is being used. For
elastic design when the third approach is being employed and the sections are
slender, then calculations inevitably are more involved as even the determination
of basic cross-sectional capacities requires allowances for local buckling effects
through the use of concepts such as the effective width technique.1
References to Chapter 13
1. Bulson P.S. (1970) The Stability of Flat Plates. Chatto and Windus, London.
2. The Steel Construction Institute (SCI) (2001) Steelwork Design Guide to BS
5950: Part 1: 2000, Vol. 1: Section Properties. Member Capacities, 6th edn. SCI,
Ascot, Berks.
Steel Designers' Manual - 6th Edition (2003)
Chapter 14
Tension members
by JOHN RIGHINIOTIS and ALAN KWAN
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
14.1 Introduction
Theoretically, the tension member transmitting a direct tension between two points
in a structure is the simplest and most efficient structural element. In many cases
this efficiency is seriously impaired by the end connections required to join tension
members to other members in the structure. In some situations (for example, in
cross-braced panels) the load in the member reverses, usually by the action of wind,
and then the member must also act as a strut. Where the load can reverse, the
designer often permits the member to buckle, with the load then being taken up by
another member.
14.2 Types of tension member
The main types of tension member, their applications and behaviour are:
(a) open and closed single rolled sections such as angles, tees, channels and the
structural hollow sections. These are the main sections used for tension
members in light trusses and lattice girders for bracing.
(b) compound sections consisting of double angles or channels. At least one axis of
symmetry is present and so the eccentricity in the end connection can be minimized. When angles or other shapes are used in this fashion, they should be
interconnected at intervals to prevent vibration, especially when moving loads
are present.
(c) heavy rolled sections and heavy compound sections of built-up H- and box sections. The built-up sections are tied together either at intervals (batten plates)
or continuously (lacing or perforated cover plates). Batten plates or lacing do
not add any load-carrying capacity to the member but they do serve to provide
rigidity and to distribute the load among the main elements. Perforated plates
can be considered as part of the tension member.
(d) bars and flats. In the sizes generally used, the stiffness of these members is very
low; they may sag under their own weight or that of workmen. Their small crosssectional dimensions also mean high slenderness values and, as a consequence,
they may tend to flutter under wind loads or vibrate under moving loads.
383
Steel Designers' Manual - 6th Edition (2003)
384
Tension members
(e) ropes and cables. Further discussion on these types of tension members is
included in section 14.7 and Chapter 5, section 5.3.
The main types of tension members are shown in Fig. 14.1.
Typical uses of tension members are:
(a) tension chords and internal ties in trusses and lattice girders in buildings and
bridges.
(b) bracing members in buildings.
(c) main cables and deck suspension cables in cable-stayed and suspension bridges.
(d) hangers in suspended structures.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Typical uses of tension members in buildings and bridges are shown in Fig. 14.2.
14.3 Design for axial tension
Rolled sections behave similarly to tensile test specimens under direct tension
(Fig. 14.1).
For a straight member subject to direct tension, F:
F
A
FL
elongation, d L =
AE
tensile stress, ft =
(in the linear elastic range)
load at yield, Py = Py A = load at failure (neglecting strain hardening)
For typical stress–strain curves for structural steel and wire rope see Fig. 14.3.
14.3.1 BS 5950: Part 1
The design of axially loaded tension members is given in Clause 4.6.1. The tension
capacity is
Pt = pyAe
where Ae is the sum of the net effective areas (defined in Clause 3.4.3). Here a steel
grade dependent factor Ke is used to determine the effective net area from the actual
net area of a member with holes, i.e. the gross area less deductions for fastener holes.
Reference should be made to clause 3.4.4.3 for members with staggered holes.
The values for coefficient Ke, given below for steels complying with BS5950-2,
come from results which show that the presence of holes does not reduce the effective capacity of a member in tension provided that the ratio of the net area to the
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
Design for axial tension
rolled sections
:i:
Fig. 14.1
Tension members
JJ
compound sections
heavy rolled and built—up sections
threaded bar
flat
round strand rope
locked coil rope
385
Steel Designers' Manual - 6th Edition (2003)
386
Tension members
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(a)
tie
hanger
braced
suspended
(b)
\_ties
1
(c)
main cables
deck
suspenders
(d)
Fig. 14.2 Tension members in buildings and bridges. (a) Single-storey building – roof and
truss bracing. (b) Multi-storey building. (c) Bridge truss. (d) Suspension bridge
Steel Designers' Manual - 6th Edition (2003)
Design for axial tension
387
2000
1756
1583
1500
t'l
E
E
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
1000
1
377
336
strain
Fig. 14.3
Stress–strain curves for structural steels and wire rope
gross area is suitably greater than the ratio of the yield strength to the ultimate
strength.
Ke = 1.2 for grade S275
Ke = 1.1 for grade S355
Ke = 1.0 for grade S460
For other steels Ke = (Us/1.2)/py, where Us is the specified minimum ultimate tensile
strength and py is the design strength.
14.3.2 BS 5400: Part 3
The member should be such that the design ultimate axial load does not exceed the
tensile resistance PD, given by (using BS 5400 notation):
PD =
s y Ae
g m g f3
Steel Designers' Manual - 6th Edition (2003)
388
Tension members
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
where Ae = effective cross-sectional area
= k1 k2 At
At = net cross-sectional area
A = gross area
sy = nominal yield stress for steel given in BS 4360
gm = partial safety factor.a Values are given in Table 2 of BS 5400: Part 3. This
takes account of material variation.
gf3 = partial safety factor.a Values are given in clause 4.3.3. This is often called
the gap factor.
k1, k2 = factors. Values are given in Clause 11.
a
Note that the design ultimate load incorporates other safety factors (gfL).
14.4 Combined bending and tension
Bending in tension members arises from
(a) eccentric connections
(b) lateral loading on members
(c) rigid frame action
When a structural member is subjected to axial tension combined with bending
about axes xx and yy, then the total stress in the section ( fmax = fr + fbx + fby) comes
from the sum of the stress from each of the separate actions: tensile (ft = F/A);
bending stress about xx axis ( fbx = Mx/Zx); and bending stress about yy axis
(fby = My/Zy). The separate stress diagrams are shown in Fig. 14.4(a). The load
values causing yield when each of the three actions act alone are:
tensile load at yield,
Py = py A
moment at yield (xx axis), Mxx = pyZx
moment at yield ( yy axis), Myy = pyZy
The values Py, Mxx and Myy form part of a three-dimensional interaction surface, and
any point on this surface gives a combination of F, Mx and My for which the
maximum stress equals the yield stress.
An elastic interaction surface for a rectangular hollow section is shown in
Fig. 14.4(b).
The maximum values for axial tension and plastic moment for the same member
are:
axial tension,
plastic moment (xx axis),
plastic moment (yy axis),
Py = pyA
Mpx = pySx
Mpy = pySy
The difference between the elastic and plastic interaction surfaces represents the
additional design strength available if plasticity is taken into account (Fig. 14.5).
Steel Designers' Manual - 6th Edition (2003)
Combined bending and tension
fF
tA 1bxZ
y
389
f
f t bx
fmox f1+ bx
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
by =
(0)
axial tension
moment
yy axis
moment
xx axis
(b)
Fig. 14.4 Combined bending and tension – elastic analysis. (a) Stress diagram. (b) Elastic
interaction surface
14.4.1 BS 5950: Part 1
The design of tension members with moments is given in Clause 4.8.2. This states
that tension members should be checked for capacity at the points of greatest
bending moments and axial loads, usually at the ends. The check can be carried out
using the ‘simplified method’ (Clause 4.8.2.2) or the ‘more exact method’. In the
simplified method, the following relationship should be satisfied:
Ft M x M y
+
+
£1
Pt Mcx Mcy
Steel Designers' Manual - 6th Edition (2003)
390
Tension members
y
xfltx
TL1JJ — tension
y
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
axial tension
M 11
Fig. 14.5
moment
moment
yy axis
xx axis
Plastic interaction surface
where Ft
Pt
Mx
Mcx
My
Mcy
= applied tensile force in the member at the critical location
= tension capacity of the member as discussed in Section 14.3.1 above
= applied moment about the major axis at the critical location
= moment capacity about the major axis in the absence of axial load
= applied moment about the minor axis at the critical location
= moment capacity about minor axis in the absence of axial load.
The moment capacity Mc for sections with low shear load is Mc = pyZeff (for slender
sections), or Mc = pyZ (for semi-compact sections), or Mc = pyS (for plastic and
compact sections), where S = plastic modulus about the relevant axis, Z = elastic
modulus about the relevant axis, and Zeff is the effective section modulus as
determined in Clause 3.6.2. In all cases, the moment capacity should be limited to
1.2pyZ.
Greater economy may be achieved for plastic or compact cross sections subject
to single plane bending through use of the ‘more exact method’ (Clause 4.8.2.3).
The bending moment is not to exceed the ‘reduced plastic moment capacity’ Mrx or
Mry for bending about the major and minor axes respectively. The bending capacities are reduced due to the presence of axial load, and the reduction comes through
a decrease in the plastic section modulus. Formulae are provided in Appendix J.2
of the Code for symmetric I- and H-sections.
The ‘more exact method’ may be used for bending in two planes only in doublysymmetric members, in which case
Ê Mx ˆ
Ë M rx ¯
z1
Ê My ˆ
+Á
˜
Ë M ry ¯
z2
£1
Steel Designers' Manual - 6th Edition (2003)
Combined bending and tension
391
where the constants z1 and z2 have values of 2.0 for solid and hollow circular
sections, 1.0 for solid and hollow rectangular sections, z1 = 2.0 and z2 = 1.0 for
symmetric I- and H-sections, and 1.0 for all other sections.
Additionally, the Code requires a member under tension and bending to
be checked for resistance to lateral-torsional buckling. In this case, the axial
force is removed, and the member is treated as a beam under bending alone (see
16.3.6).
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
14.4.2 BS 5400: Part 3
A member subjected to co-existent tension and bending should be such that at all
cross sections the following relationship is satisfied:
My
P
Mx
+
+
£ 1.0
PD MDxt MDyt
= axial tensile force
= tensile resistance, see section 14.3.2
= bending moments at the section about xx and yy axes,
respectively
MDxt, MDyt = corresponding bending resistances with respect to the extreme
tensile fibres.
where P
PD
Mx, My
Additionally, if at any section within the middle third of the length of the member
the maximum compressive stress due to bending exceeds the tensile stress due to
axial load, the design should be such that:
M x max M y max
P
+
< 1+
MDxc
MDyc
PD
where Mxmax, Mymax = maximum moments anywhere within the middle third
MDxc, MDyc = corresponding bending resistances with respect to the extreme
compression fibres.
For compact sections the bending resistance is given in clause 9.9.1.2. This is:
MD =
Zpes Ic
g mg f3
where sIc = limiting compressive stress (see clause 9.8)
Zpc = plastic modulus of the effective section (see clause 9.4.2).
Clause 9.9.1.3 should be referred to for the resistance of non-compact sections.
Steel Designers' Manual - 6th Edition (2003)
392
Tension members
14.5 Eccentricity of end connections
Simplified design rules are given in both BS 5950 and BS 5400 for the effects of
combined tension and bending caused solely by the eccentric load introduced into
the member by the end connection.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
14.5.1 BS 5950: Part 1
BS 5950 provides simplified design rules for the common single- and double-angles,
channels and T-sections. Eccentrically connected tension members may still be
designed as axially loaded members but with a reduction in the cross sectional area.
Capacities of single angles connected through one leg, single channels connected
only through the web and T-sections connected only through the flange, are given
by
for bolted connections,
Pt = py ( Ae - 0.5a2 )
for welded connections, Pt = py ( Ag - 0.3a2 )
where a2 is the gross cross-sectional area (Ag) less the gross area of the connected
element (= thickness ¥ overall leg width for an angle, the overall depth for a channel,
or the flange width for a T-section).
Capacities of back-to-back double angles connected through one leg only, double
channels connected through the web, or double T-sections connected through the
flange only, are given by
for bolted connections,
Pt = py ( Ae - 0.25a2 )
for welded connections, Pt = py ( Ag - 0.15a2 )
provided that the two component sections are sufficiently connected along their
lengths so as to enable them to act compositely (i.e. connected to both sides of a
gusset or section, and the components are interconnected by bolts or welds and held
apart, by battens or solid packing pieces, in at least two locations within their length).
Where the two component sections are both connected to the same side of a gusset
or section, or insufficiently interconnected along their lengths, then they should
be treated as single sections and their individual capacity determined as in the
previous paragraph.
14.5.2 BS 5400: Part 3
The bending moment resulting from an eccentricity of the end connections should
be taken into account. For single angles connected by one leg these provisions may
be considered to be met if the effective area of the unconnected leg is taken as:
Steel Designers' Manual - 6th Edition (2003)
Other considerations
393
Ê 3 A1 ˆ
A
Ë 3 A1 + A2 ¯ 2
where A1 = net area of the connected leg
A2 = net area of the unconnected leg.
14.6 Other considerations
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
14.6.1 Serviceability, fatigue and corrosion
Ropes and bars are not normally used in building construction because they lack
stiffness, but they have been used in some cases as hangers in suspended buildings.
Very light, thin tension members are susceptible to excessive elongation under
direct load as well as lateral deflection under self-weight and lateral loads. Special
problems may arise where the members are subjected to vibration or conditions
leading to failure by fatigue, such as can occur in bridge deck hangers. Damage
through corrosion is undesirable; adequate protective measures must be adopted.
All these factors can make the design of tension members a complicated process in
some cases.
The light-rolled sections used for tension members in trusses and for bracing are
easily damaged during transport. It is customary to specify a minimum size for such
members to prevent this happening. For angle ties, a general rule is to make the leg
length not less than one-sixtieth of the member length.
Ties subject to load reversal, e.g. under the action of wind, could buckle. Where
such buckling is dangerous (e.g. due to a lack of alternative load paths) or merely
unsightly, tension members should also be checked as a compressive member
(see 15.3).
In a fatigue assessment of a member, the stress range considered is the greatest
algebraic difference between principal stresses occurring during any one stress cycle.
The above applies when the stresses occur on principal planes of more than 45°
apart.
The stresses should be calculated using elastic theory and taking account of all
axial, bending and shear stresses occurring under the design loadings. Alternatively
the design loadings in BS 5400: Part 10 can be used.
In assessing the fatigue behaviour of a member, the following effects have to be
included:
•
•
•
•
shear lag, restrained torsion and distortion, transverse stresses and flange
curvatures,
effective width of steel plates,
stresses in triangulated skeletal structures due to load applications away from the
joints, member eccentricities to joints and rigidity of joints,
cracking of concrete in composite elements.
Steel Designers' Manual - 6th Edition (2003)
394
Tension members
The following effects however, can be ignored:
•
•
•
residual stresses,
eccentricities unnecessarily arising in a standard detail,
plate buckling.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
14.6.2 Stress concentration factors
In cases of geometrical discontinuity, such as a change of cross section or an aperture, the resulting stress concentrations may be determined either by special analysis or by the use of stress concentration factors. Stress concentrations are not usually
important in ductile materials but can be the cause of failure due to fatigue or brittle
fracture in certain considerations. A hole in a flat member can increase the stresses
locally on the net section by a factor which depends on the ratio of hole diameter
to net plate width.
When designing against fatigue it is convenient to consider three levels of stress
concentration:
(a) stress concentrations from structural action due to the difference between the
actual structural behaviour and the static model chosen,
(b) macroscopic stress concentrations due to large scale geometric interruptions to
stress flow,
(c) microscopic and local geometry stress concentrations due to imperfections
within the weld or the heat affected zone.
For the design of welded details and connections further reference should be
made to BS 5400: Part 10. However, for non-standard situations it may be necessary to determine the stress concentration factor directly from a numerical analysis
study or from an experimental model.
In many cases the detail under consideration is very likely not to fit neatly into
one of the classes. On site, the actual stress range for a particular loading occurrence
is likely to be strongly influenced by detailed fit of the joint and overall fit of the
structure. Therefore the overall form should be such that load paths are as smooth
as possible, and unintended load paths should be avoided, particularly where fit
could significantly influence behaviour. Discontinuities must be avoided by tapering and appropriate choice of radii.
14.6.3 Fabrication and erection
The behaviour of tension members in service depends on the fabrication tolerances
and the erection sequence and procedure. Care must be taken to ensure that no
Steel Designers' Manual - 6th Edition (2003)
Cables
395
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
tension member is slack after erection, so that they are all immediately active in
resisting service loads.
Screwed ends and turnbuckles can be used to adjust lengths of bars and cables
after they are in place. Bracing members fabricated from rolled sections should be
installed and properly tightened before other connections and column base plates
are bolted up, to bring the structure into line and square. Bracing members are
usually specified slightly shorter than the exact length to avoid sagging and allow
them to be immediately effective.
Complete or partial trial shop assembly is often specified in heavy industrial
trusses and bridge members to ensure that the fabrication is accurate and that
erection is free from problems.
14.7 Cables
This section is directed mainly towards bridge cables; see Chapter 5, section 5.3 for
cables in building structures.
14.7.1 Composition
A cable may be composed of one or more structural ropes, structural strands, locked
coil strands or parallel wire strands. A strand, with the exception of a parallel wire
strand, is an assembly of wires formed vertically around a central wire in one
or more symmetrical layers. A strand may be used either as an individual loadcarrying member, where radius of curvature is not a major requirement, or as a
component in the manufacture of structural rope.
A rope is composed of a plurality of strands vertically laid around a core. In
contrast to the strand, a rope provides increased curvature capability and is used
where curvature of the cable becomes an important consideration. The significant
differences between strand and rope are as follows:
•
•
•
•
at equal sizes, a rope has lower breaking strength than a strand,
the modulus of elasticity of a rope is lower than that of a strand,
a rope has more curvature capability than a strand,
the wires in a rope are smaller than those in a strand of the same diameter:
consequently, a rope for a given size coating is less corrosion-resistant because
of the thinner coating on the smaller diameter wires.
14.7.2 Application
Cables used in structural applications, namely for suspension systems in bridges, fall
into the following categories:
Steel Designers' Manual - 6th Edition (2003)
396
(a)
(b)
(c)
(d)
Tension members
parallel-bar cables;
parallel-wire cables;
stranded cables (see Fig. 14.1);
locked-coil cables (see Fig. 14.1).
The final choice depends on the properties required by the designer, i.e. modulus
of elasticity, ultimate tensile strength, durability. Other criteria include economic
and structural detailing, i.e. anchorages, erection, etc.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
14.7.3 Parallel-bar cables
Parallel-bar cables are formed of steel rods or bars, parallel to each other in metal
ducts, kept in position by polyethylene spacers. The process of tensioning the bar or
rods individually is simplified by the capability of the bars to slide longitudinally.
Cement grout, injected after erection, makes sure that the duct plays its part in
resisting the stresses due to live loads.
Transportation in reels is only possible for the smaller diameters while for the
larger sizes delivery is made in straight bars 15.0–20.0 m in length. Continuity of
the bars has to be provided by the use of couplers, which considerably reduces the
fatigue strength of the stay.
The use of mild steel necessitates larger sections than when using high-strength
wires or strands. This leads to a reduction in the stress variation and thus lessens
the risk of fatigue failure.
14.7.4 Parallel-wire cables
Parallel wires are used for cable-stayed bridges and pre-stressed concrete. Their
fatigue strength is satisfactory, mainly because of their good mechanical properties.
14.7.5 Corrosion protection
Wires in the cables should be protected from corrosion. The most effective protection is obtained by hot galvanizing by steeping or immersing the wires in a bath of
melted zinc, automatically controlled to avoid overheating. A wire is described as
terminally galvanized or galvanized re-drawn depending on whether the operation
has taken place after drawing or in between two wire drawings prior to the wire
being brought to the required diameter. For reinforcing bars and cables, the first
method is generally adopted. A quantity of zinc in the range of 250–330 g/m2 is
deposited, providing a protective coating 25–45 mm thick.
Steel Designers' Manual - 6th Edition (2003)
Further reading
397
14.7.6 Coating
The coating process, used currently for locked-coil cables, consists of coating
the bare wires with an anti-corrosion product with a good bond and long service
life. The various substances used generally have a high dropping point so as not to
run back towards the lower anchorages. They are usually high viscosity resins or
oil-based grease, paraffins or chemical compounds.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
14.7.7 Protection of anchorages
The details of the connections between the ducts and the anchorages must prevent
any inflow or accumulation of water. The actual details depend on the type of
anchorages used, on the protective systems for the cables, and on their slope. There
are different arrangements intended to ensure water tightness of vital zones.
14.7.8 Protection against accidents
Cables should be protected against various risks of accident, such as vehicle impact,
fire, explosion and vandalism. Measures to be taken may be based on the following:
(a) protection of the lower part of the stay, over a height of about 2.0 m, by a steel
tube fixed into the deck and fixed into the duct; the tube dimensions (thickness
and diameter) must be adequate.
(b) strength of the lower anchorage against vehicle impact.
(c) replacement of protective elements possible without affecting the cables themselves and, as far as possible, without interrupting traffic.
Further reading for Chapter 14
Adams P.F., Krentz H.A. & Kulak G.L. (1973) Canadian Structural Steel Design.
Canadian Institute of Steel Construction, Ontario.
Bresler B., Lin T.Y. & Scalzi J.B. (1968) Design of Steel Structures. Wiley, Chichester.
Dowling P.J., Knowles P. & Owens G.W. (1988) Structural Steel Design. Butterworths,
London.
Horne M.R. (1971) Plastic Theory of Structures, 1st edn. Nelson, Walton-on-Thames.
Owens G.W. & Cheal B.D. (1989) Structural Steelwork Connections. Butterworths,
London.
Timoshenko S.P. & Goodier J.N. (1970) Theory of Elasticity. McGraw-Hill,
London.
Steel Designers' Manual - 6th Edition (2003)
398
Tension members
Toy M. (1995) Tensile Structures. Academy Editions, London.
Trahair N.S. (2001) The Behaviour and Design of Steel Structures to BS 5950,
3rd edn. Spon Press, London.
Troitsky M.S. (1988) Cable-Stayed Bridges, 2nd edn. BSP Professional Books,
London.
Vandenberg M. (1988) Cable Nets. Academy Editions, London.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
A series of worked examples follows which are relevant to Chapter 14.
Steel Designers' Manual - 6th Edition (2003)
Worked examples
Subject
The
Steel Construction
Institute
Chapter ref.
TENSION MEMBERS
14
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950
Made by
ASKK
Checked by
Sheet no.
1
BD
BS 5950:
Part 1
Problem 1
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Design a single angle tie for the member AB shown.
At
Tensile force in member AB
Dead load
Imposed load
Total
= 122 kN
= 220kN
= 342kN
Material:
Use Steel Grade S355
Connections:
a) Welded
b) Bolted
Factored load
F = (1.4 ¥ 122) + (1.6 ¥ 220 ) = 523 kN
Table 2
Material Steel Grade S355, thickness £ 16 mm
Design strength py = 355 N/mm2
Table 9
a) Welded connections
Try 125 ¥ 75 ¥ 10 connected by long leg. Code allows angle section with
eccentric loading to be treated with a reduced tension capacity as given
in Clause 4.6.3.
x 75x IOL
120
f
4.6.2
399
Steel Designers' Manual - 6th Edition (2003)
Worked examples
400
Subject
The
Steel Construction
Institute
Chapter ref.
TENSION MEMBERS
14
Silwood Park, Ascot, Berks SL5 7QN
Made by
Design code
BS 5950
ASKK
Checked by
Sheet no.
2
BD
4.6.3.1
Area for reduced tension capacity
2
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
a1 = 120 ¥ 10 = 1200 mm
Gross area Ag = 1910 mm2
a2 = Ag - a1 = 1910 - 1200 = 710 mm2
Tension capacity Pt = py(Ag - 0.3a2)
4.6.3.1
Pt = 355 (1900 - 0.3 ¥ 710 ) = 598885 N ª 599 kN > 523 kN \ OK
\ Use 125 ¥ 75 ¥ 10 Steel Grade S355
b) Bolted connections
Try 150 ¥ 75 ¥ 10 connected by long leg. Code allows angle section with
eccentric loading to be treated with a reduced tension capacity as given in
Clause 4.6.3.
4.6.2
-'lh'°
/50 x 75 x /0 L
4mrn lw/es for 22 mm dia. HSFG bolts
70]I
I Gusset
145
Area for reduced tension capacity
4.6.3.1
a1 = 120 ¥ 10 = 1200 mm2
Effective net area Ae = sum of effective net areas ae
Here, there is only one element, with net effective area
ae = Kean
4.6.1
3.4.3
For steel grade S355, Ke = 1.1.
Hence
an = Gross area - bolt hole = 2160 - 24 ¥ 10 = 1920 mm2
ae = 1.1 ¥ 1920 = 2112 mm2 (Check that ae £ ag)
Ae = 2112 mm2
a2 = Ag - a1 = 2160 - 1200 = 960 mm2
Tension capacity Pt = py(Ae - 0.5a2)
Pt = 355 ( 2112 - 0.5 ¥ 960 ) = 579360 N ª 579 kN > 523 kN \ OK
\ Use 150 ¥ 75 ¥ 10 Steel Grade S355
3.4.3
4.6.3.1
Steel Designers' Manual - 6th Edition (2003)
Worked examples
Subject
The
Steel Construction
Institute
401
Chapter ref.
TENSION MEMBERS
14
Silwood Park, Ascot, Berks SL5 7QN
Made by
Design code
BS 5950
ASKK
Checked by
Sheet no.
3
BD
Problem 2
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Design a double angle tie for the member AB in Problem 1. Assume the use of double
angles, connected back to back through a 8 mm gusset. Assume welded connections.
Code allows angle section with eccentric loading to be treated
with a reduced tension capacity as given in Clause 4.6.3.
4.6.2
Try 2 ¥ (75 ¥ 50 ¥ 8), connected through long leg.
Effective Area, Ae
a1 =71 x 8 =568mm2
a2 =46 x 8 =368 main2
2x
Ae
Ae =
=
2
5xa1 1 xa2l
j
[ai÷[ 5xaj÷a2J
I
x [568
5x568
[5 x 568
1
368 J
x368
I
1788 mm2
Area for reduced tension capacity
4.6.3.2
a1 = 2 ¥ (75 ¥ 8 ) = 1200 mm
Gross area Ag = 2 ¥ 941 = 1882 mm2
a2 = Ag - a1 = 1882 - 1200 = 682 mm2
2
Tension capacity Pt = py(Ag - 0.15a2)
4.6.3.2
Pt = 355 (1882 - 0.15 ¥ 682) = 631793 N ª 632 kN > 523 kN \ OK
\ Use 2 ¥ (75 ¥ 50 ¥ 8) Steel Grade S355
Note: The two angles must be connected at regular intervals.
4.6.3.2
Steel Designers' Manual - 6th Edition (2003)
Chapter 15
Columns and struts
by DAVID NETHERCOT
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
15.1 Introduction
Members subject to compression, referred to as either ‘columns’ or ‘struts’, form
one of the basic types of load-carrying component. They may be found, for example,
as vertical columns in building frames, in the compression chords of a bridge truss
or in any position in a space frame.
In many practical situations struts are not subject solely to compression but,
depending upon the exact nature of the load path through the structure, are also
required to resist some degree of bending. For example, a corner column in a building is normally bent about both axes by the action of the beam loads, a strut in a
space frame is not necessarily loaded concentrically, the compression chord of a roof
truss may also be required to carry some lateral loads. Thus many compression
members are actually designed for combined loading as beam-columns. Notwithstanding this, the ability to determine the compressive resistance of members is of
fundamental importance in design, both for the struts loaded only in compression
and as one component in the interaction type of approach normally used for beamcolumn design.
The most significant factor that must be considered in the design of struts is buckling. Depending on the type of member and the particular application under consideration, this may take several forms. One of these, local buckling of individual
plate elements in compression, has already been considered in Chapter 13. Much of
this chapter is devoted to the consideration of the way in which buckling is handled
in strut design.
15.2 Common types of member
Various types of steel section may be used as struts to resist compressive loads;
Fig. 15.1 illustrates a number of them. Practical considerations such as the methods
to be employed for making connections often influence the choice, especially for
light members. Although closed sections such as tubes are theoretically the most
efficient, it is normally much easier to make simple site connections, using the
minimum of skilled labour or special equipment, to open sections. Typical arrangements include:
402
Steel Designers' Manual - 6th Edition (2003)
Design considerations
0 Li
\\\\\\
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(a)
(b)
TI
(1)
Fig. 15.1
(g)
(c)
403
_ JL
(d)
(e)
__
(h)
(I)
Typical column cross sections
(1) light trusses and bracing – angles (including compound angles back to back)
and tees
(2) larger trusses – circular hollow sections, rectangular hollow sections, compound
sections and universal columns
(3) frames – universal columns, fabricated sections e.g. reinforced UCs
(4) bridges – box columns
(5) power stations – stiffened box columns.
15.3 Design considerations
The most important property of a strut as far as the determination of its loadcarrying capacity is concerned is its slenderness, l, defined as the ratio of its effective length, LE, divided by the appropriate radius of gyration, r. Codes of practice
such as BS 5950 used to place upper limits on l, as indicated in Table 15.1, so as to
avoid the use of flimsy construction, i.e. to ensure that a member which will ordinarily be subject only to axial load does have some limited resistance to an accidental lateral load, does not rattle, etc. Although not now explicitly stated in the
2000 version of the code, it is still good practice to aim for robust construction.
Similarly, for BS 5400: Part 3, no actual limits are specified but the user is provided
with a general advisory note to the effect that construction should be suitably
robust. By contrast, BS 5400 does place upper bounds on local plate slenderness to
avoid consideration of local buckling (see Table 15.2).
Strut design will normally require that, once a trial member has been selected and
its loading and support conditions determined, attention be given to whichever of
the following checks are relevant for the particular application:
Steel Designers' Manual - 6th Edition (2003)
404
Columns and struts
Table 15.1 Maximum slenderness values for struts
Condition
BS 5950: Part 1 limits
Members in general
Members resisting self-weight wind loads only
Members normally in tension but subject to load
reversal due to wind
a
BS 5950: Part 5
a
180
250a
–
180
250
350
Check for self-weight deflection if l > 180; allow for bending effects in design if this deflection > L/1000
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Table 15.2 Upper limits on plate slenderness of BS 5400: Part 3
b/t or D/t limita
Unstiffened
outstand
Stiffened
outstand
CHS
py = 355 N/mm2
py = 275 N/mm2
12
14
14
16
100
114
a
See Fig. 13.1 for definitions of b; D = outside diameter
(1) Overall flexural buckling – largely controlled by the slenderness ratio, l, which
is a function of member length, cross-sectional shape and the support conditions provided; also influenced by the type of member
(2) Local buckling – controlled by the width-to-thickness ratios of the component
plate elements (see Chapter 13); with some care in the original choice of
member this need not involve any actual calculation
(3) Buckling of component parts – only relevant for built-up sections such as laced
and battened columns; the strength of individual parts must be checked, often
by simply limiting distances between points of interconnection
(4) Torsional or torsional-flexural buckling – for cold-formed sections and in
extreme cases of unusually shaped heavier open sections, the inherent low
torsional stiffness of the member may make this form of buckling more critical
than simple flexural buckling.
In principle, local buckling and overall buckling (flexural or torsional) should always
be checked. In practice, provided cross sections that at least meet the semi-compact
limits for pure compression are used, then no local buckling check is necessary since
the cross section will be fully effective.
15.4 Cross-sectional considerations
Since the maximum attainable load-carrying capacity for any structural member is
controlled by its local cross-sectional capacity (factors such as buckling may prevent
this being achieved in practice), the first step in strut design must involve considera-
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Cross-sectional considerations
405
tion of local buckling as it influences axial capacity. Only two classes of section are
relevant for purely axially compressed members: either the section is not slender,
in which case its full capacity, py A, is available, or it is slender and some allowance
in terms of a reduced capacity is required. The distinctions between plastic, compact
and semi-compact as described in Chapter 13 therefore have no relevance when the
type of member under consideration is a strut.
The general approach for a member containing slender plate elements designed
in accordance with BS 5950: Part l utilizes an effective cross-section as illustrated
in Fig. 15.2. Essentially the effective portion of any slender plate element is taken
as the maximum width of that class of element that corresponds to the semi-compact
limit. A slightly simpler, but generally much more conservative option requires the
use in all calculations relating to that member, apart from those concerned with
components of connections to the member, of a reduced design strength, with the
magnitude of the reduction being dependent on the extent to which the semicompact limits (the boundary between not slender and slender) are exceeded.
I
I
2Otx
I
I
l5Tc l5Tx
///A
2Otc
1___
'F
F///l
l3Tc l3Tx
l3Tx l3Tx
2Otc
20t
-I
I
rolled I-section
hot finished RHS
Fig. 15.2
15Th iSTs
I
/'
r/z/,1
I
welded I-section
cold finished RHS
roIled H-section
welded H-section
welded box section
Effective cross-sections under pure compression. (Based on Figure 8(a) in BS 5950:
2000-1)
Steel Designers' Manual - 6th Edition (2003)
406
Columns and struts
Effectively this means that the member is assumed to be made of a steel having a
lower material strength than is actually the case. As noted above, BS 5400: Part 3
does not permit the use of columns with slender outstands.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
15.4.1 Columns with slender webs
Reference 1 indicates that most UB and a few RHS, even in S275 steel, will be
slender according to BS 5950: Part 1 if used as struts. A suitable allowance for the
weakening effect of local buckling should therefore be made. Two different procedures are given, outlined in the following example.
Consider a 406 ¥ 140 ¥ 39 UB (the UB with the highest d/t) in S275 material. This
has a d/t of 57.1 compared with the limit in Table 11 of 40 assuming the section to
be fully stressed: i.e. r2 as defined in clause 3.5.5 is equal to unity (for r2 < 1 a higher
limit is applicable).
(1) Use clause 3.6.5, limiting d/t = 40(275/pyr)0.5
2
and reduced strength, pyr = (40 57.1) ¥ 275
= 134.8 N/mm 2
For LE/ry = 100, using Appendix C for pyr = 134.8 N/mm2 and strut curve ‘a’
pc = 103.5 N/mm 2
Pc = 49.4 ¥ 103.5 10
= 511 kN
(2) Since the web is slender, obtain Aeff from clause 3.6.2.2 using Figure 8 of
BS 5950: 2001-1.
Aeff = 4970 - 40 ¥ 6.4
= 4714 mm2
For sections containing class 4 element(s) clause 4.7.4 gives
Pc = Aeff pcs
in which pcs is obtained for a slenderness l(Aeff/Ag)0.5
= 100(4714 4970)
0.5
= 97.0
From Table 24, using strut curve ‘a’ and py = 275 N/mm2, pcs = 164 N/mm2
\ Pc = 4714 ¥ 164 1000
= 773 kN
This compares with the value for a ‘fully effective equivalent’ of 780 kN.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Cross-sectional considerations
407
The second method leads to a significantly higher result, being only some 1%
below the fully effective figure. The main reason for this is that the first method
applies the same reduction in strength, based on the proportions of the most slender
plate element, to the whole of the cross section, whereas the second method makes
a reduction only for the plate elements which are slender.
Both BS 5400: Part 3 and BS 5950: Part 5 employ a method for slender sections
in compression that uses the concept of effective area. Its implementation has been
illustrated in general terms by the second method of the above example. The bridge
code prohibits the use of slender outstands by specifying the absolute upper limits
on b/t given in Table 15.2 with no procedure to cover arrangements outside these,
and determining the effective area of the cross section by summing the full areas of
any outstands and compact webs and the effective areas, KcAc, of the slender webs.
Values of Kc are obtained as a function of b/t from a graph.
It is worth emphasizing that for design using hot-rolled sections the majority of
situations may be treated simply by ensuring that the proportions of the cross
section lie within the semi-compact limits. Reference 1, in addition to listing flange
b/T and web d/t values for all standard sections, also identifies those sections that
are slender according to BS 5950: Part 1 when used in either S275 or S355 steel.
Table 15.3 lists these.
Table 15.3 Sections classified as ‘slender’ when used as struts
Section
Universal
beam1
S275
All except
1016 ¥ 305 ¥ 487
1016 ¥ 305 ¥ 437
1016 ¥ 305 ¥ 393
914 ¥ 419 ¥ 388
610 ¥ 305 ¥ 238
610 ¥ 305 ¥ 179
533 ¥ 210 ¥ 122
457 ¥ 191 ¥ 98
457 ¥ 191 ¥ 89
457 ¥ 152 ¥ 82
406 ¥ 178 ¥ 74
356 ¥ 171 ¥ 67
356 ¥ 171 ¥ 57
305 ¥ 165 ¥ 54
305 ¥ 165 ¥ 46
305 ¥ 127 ¥ 48
305 ¥ 127 ¥ 42
305 ¥ 127 ¥ 37
254 ¥ 146 ¥ 43
254 ¥ 146 ¥ 37
254 ¥ 146 ¥ 31
254 ¥ 102 ¥ 28
254 ¥ 102 ¥ 25
254 ¥ 102 ¥ 22
203 ¥ 133 ¥ 30
203 ¥ 133 ¥ 25
203 ¥ 102 ¥ 23
178 ¥ 102 ¥ 19
S355
All except
1016 ¥ 305 ¥ 487
1016 ¥ 305 ¥ 437
1016 ¥ 305 ¥ 393
610 ¥ 305 ¥ 238
356 ¥ 171 ¥ 67
305 ¥ 165 ¥ 54
305 ¥ 127 ¥ 48
305 ¥ 127 ¥ 42
254 ¥ 146 ¥ 43
254 ¥ 146 ¥ 37
203
203
203
178
¥
¥
¥
¥
133
133
102
102
¥
¥
¥
¥
30
25
23
19
Steel Designers' Manual - 6th Edition (2003)
408
Columns and struts
Table 15.3 Continued
Section
S275
S355
152 ¥ 89 ¥ 16
127 ¥ 76 ¥ 13
Universal
column
none
none
Circular
hollow
section1,2
508 ¥ 8.0
406.4 ¥ 6.3
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Rectangular
hollow
section1,2
250
300
400
450
500
Square
hollow
sections1,2
Rolled
steel
angles
152 ¥ 89 ¥ 16
127 ¥ 76 ¥ 13
¥
¥
¥
¥
¥
150
200
200
250
300
¥
¥
¥
¥
¥
5
6.3
8
8 and 10
8 and 10
300 ¥ 300 ¥ 6.3
350 ¥ 350 ¥ 8
200 ¥ 150 ¥ 12
200 ¥ 100 ¥ 12
200 ¥ 100 ¥ 10
125 ¥ 75 ¥ 8
100 ¥ 50 ¥ 6
200 ¥ 200 ¥ 16
150 ¥ 150 ¥ 12
150 ¥ 150 ¥ 10
120 ¥ 120 ¥ 8
100 ¥ 100 ¥ 8
90 ¥ 90 ¥ 7
75 ¥ 75 ¥ 6
50 ¥ 50 ¥ 4
160
200
200
250
300
400
450
500
¥
¥
¥
¥
¥
¥
¥
¥
80 ¥ 4
100 ¥ 5
120 ¥ 5
150 ¥ 5 and 6.3
200 ¥ 6.3
200 ¥ 8 and 10
250 ¥ 8 and 10
300 ¥ 8 to 12.5
200
250
300
350
400
¥
¥
¥
¥
¥
200
250
300
350
400
¥
¥
¥
¥
¥
5
6.3
6.3
8
10
200 ¥ 150 ¥ 15
200 ¥ 150 ¥ 12
200 ¥ 100 ¥ 15
200 ¥ 100 ¥ 12
200 ¥ 100 ¥ 10
150 ¥ 90 ¥ 10
150 ¥ 75 ¥ 10
125 ¥ 75 ¥ 8
100 ¥ 75 ¥ 8
100 ¥ 65 ¥ 7
80 ¥ 40 ¥ 6
65 ¥ 50 ¥ 5
200 ¥ 200 ¥ 18
200 ¥ 200 ¥ 16
150 ¥ 150 ¥ 12
150 ¥ 150 ¥ 10
120 ¥ 120 ¥ 10
120 ¥ 120 ¥ 8
100 ¥ 100 ¥ 8
90 ¥ 90 ¥ 8
90 ¥ 90 ¥ 7
75 ¥ 75 ¥ 6
70 ¥ 70 ¥ 6
60 ¥ 60 ¥ 5
50 ¥ 50 ¥ 4
1
BS 5950-1: 2000 clause 3.5.5 presents equations for calculating
stress ratios r1 and r2, which are used for the classification of the web
of I-, H- and box sections in Table 11 and RHS in Table 12. These
ratios allow the influence of the applied axial force on the local buckling resistance of the section to be taken into account during the
section classification process. The cross-sections noted as slender in
this table are assumed to be fully loaded in compression.
2
Hot finished tubular steel to BS EN 10210-2: 1997
Steel Designers' Manual - 6th Edition (2003)
Compressive resistance
409
15.5 Compressive resistance
The axial load-carrying capacity for a single compression member is a function of
its slenderness, its material strength, cross-sectional shape and method of manufacture. Using BS 5950: Part 1, the compression resistance, Pc, is given by clause 4.7.4
as
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Pc = Ag pc
in which Ag is the gross area and pc is the compressive strength.
Values of pc in terms of slenderness l and material design strength py are given
in Table 24. Slenderness is defined as the ratio of the effective length, LE (taken as
the geometrical length L for the present but see section 15.7), to the least radius of
gyration, r.
The basis for Table 24 is the set of four column curves shown in Fig. 15.3. These
have resulted from a comprehensive series of full-scale tests, supported by detailed
numerical studies, on a representative range of cross sections.2 They are often
referred to as the European Column Curves. Four curves are used in recognition of
the fact that for the same slenderness certain types of cross section consistently
perform better than others as struts. This is largely due to the arrangement of the
material but is also influenced by the residual stresses that form as a result of differential cooling after hot rolling. It is catered for in design by using the strut curve
selection table given as Table 23 in BS 5950: Part 1.
The first step in column design is therefore to consult Table 23 to see which strut
curve of Table 24 is appropriate. For example, if the case being checked is a UC
liable to buckle about its minor axis, Table 24 for strut curve c should be used. Selection of a trial section fixes r and Ag; the geometrical length will be defined by the
application required, so l and thus pc and Pc may be obtained.
a)
• b)
• c)
•d)
0
slenderness X
Fig. 15.3
Strut curves of BS 5950
= 0.002
= 0.0035
= 0.0055
= 0.008
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
410
Columns and struts
The above process should be used for all types of rolled section, including those
reinforced by the addition of welded cover plates for which Fig. 14 should be used
to supplement Table 23 when deciding which curve of Table 24 to use for determining pc. When sections are fabricated by welding plates together, however, the
pattern of residual stresses produced by the heating and cooling of the welding
process will be rather different from that typically found in a hot-rolled section3 as
illustrated in Fig. 15.4. This tends to produce lower buckling strengths. It is allowed
for in BS 5950: Part 1 by the modification to the column curves shown in Fig. 15.5
in which the basic curve of Fig. 15.3 is replotted for a reduced material design
strength. This has the effect of reducing the basic curve down to the level of that
for the welded member in the important medium slenderness region. Clause 4.7.5
thus requires that when checking the axial resistance of a member that has been
fabricated from plate by welding, the value of pc be obtained using a py value of
20 N/mm2 less than the actual figure.
BS 5400: Part 3 is intended to cover a wider range of construction than BS 5950:
Part 1. Specifically in this context it recognizes the greater likelihood that struts may
contain slender elements (other than outstands) and so it defines compression resistance, Pc, as
—
,'T •'\
I
,ji
1V
II
ill
lF\
III
\I,
III
C
'7
Ill
ill
Ill
I
II /
1'
___ ___
C. Al C
I
rolled universal beam
rolled universal column
I
II
I
I
I
I
I
1
—
I
I
'
C
')
,-
T
.
welded box
Fig. 15.4
Distribution of residual stresses
I
,/
Steel Designers' Manual - 6th Edition (2003)
Torsional and flexural-torsional buckling
411
actual weld
20 N/mm2 reduced curve
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
slenderness X
Fig. 15.5
Modified strut curve for welded sections used in BS 5950: Part 1
Pc = Aepc
(clause 10.6.1.1)
in which Ae is the effective area defined as
Ae = SKc (kh Ac )
for members other than CHSs
where Kc allows for loss of effectiveness in slender plate elements, determined from
Figure 36, kh allows for the presence of holes, and Ac is the net area; and
Ae = Ac
for CHSs for which
D Ê py ˆ ˘
È
Ae = Ac Í1.15 - 0.003
Á
˜
t Ë 355 ¯ ˙˚
Î
D Ê py ˆ
Á
˜ £ 50
t Ë 335 ¯
for CHSs for which
D Ê py ˆ
Á
˜ > 50
t Ë 355 ¯
pc is obtained from a set of curves (Figure 37) similar to Fig. 15.3.
Selection of the appropriate column curve is made using a simpler selection table
than that of Part 1 of BS 5950. Essentially it distinguishes between curves on the
basis of the ratio of radius of gyration to distance from neutral axis to extreme fibre,
apart from heavy sections and hot-finished SHS, which are universally allocated to
the lowest and highest curves respectively.
As noted in the example of 15.4.1, for class 4 sections the slenderness l may be
reduced in the ratio of the square root of the effective to the gross area.
15.6 Torsional and flexural-torsional buckling
In addition to the simple flexural buckling described in the previous section, struts
may buckle due to either pure twisting about their longitudinal axis or a combination of bending and twisting. The first type of behaviour is only possible for
centrally-loaded doubly-symmetrical cross sections for which the centroid and shear
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
412
Columns and struts
centre coincide. The second, rather more general, form of response occurs for
centrally-loaded struts such as channels for which the centroid and shear centre do
not coincide.
In practice pure torsional buckling of hot-rolled structural sections is highly
unlikely, the pure flexural mode normally requiring a lower load, unless the strut is
of a somewhat unusual shape so that its torsional and warping stiffnesses are low
as for a cruciform section. In such cases a reasonable design approach consists of
determining an effective slenderness based on the direct use of the member’s elastic
critical load for torsional buckling (assuming this to be lower than its elastic critical load for pure flexural buckling) and using this to enter the basic column design
curve. This approach is well substantiated for aluminium members for which torsional buckling more commonly controls. Similarly, for unsymmetrical sections use
of an effective slenderness based on the member’s lowest elastic critical load, corresponding to flexural-torsional buckling in this case, permits the basic column
design curve to be retained and used in a more general way. For certain types of
section, such as hot-rolled angles, special empirically-based design approaches are
provided in BS 5950: Part 1 which recognize the possibility of some torsional influence. They are discussed in detail in section 15.8.
Torsional-flexural buckling is of greater practical significance in the design and
use of cold-formed sections. This arises for two reasons:
(1) Because the torsion constant, J, depends on t3, the use of thin material results
in the ratio of torsional to flexural stiffness being much reduced as compared
with hot-rolled sections
(2) The forming process leads naturally to a preponderance of singly-symmetrical
or unsymmetrical open sections as these can be produced from a single sheet.
Procedures are given in BS 5950: Part 5 for determining the axial strength of
singly-symmetrical sections using a factored effective length aLE, in which a is
obtained from
a =1
a = PEY PTF
for PEY £ PTF
for PEY > PTF
Formulae for PTF in terms of basic cross-sectional properties are also provided. The
use of a π 1 is only required for those situations illustrated in Fig. 15.6 for which
PTF is the lowest elastic critical load.
15.7 Effective lengths
Basic design information relating column strength to slenderness is normally
founded on the concept of a pin-ended member, e.g. Fig. 15.3. Stated more precisely,
this means a member whose ends are supported such that they cannot translate
Steel Designers' Manual - 6th Edition (2003)
Effective lengths
413
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
load
Fig. 15.6
Elastic critical load curves for a member subject to flexural–torsional buckling
relative to one another but are able to rotate freely. Compression members in actual
structures are provided with a variety of different support conditions which are
likely to be less restrictive in terms of translational restraint, giving fixity in position, with or without more restriction in terms of rotational restraint, giving fixity
in direction.
The usual way of treating this topic in design is to use the concept of an effective
column length, which may be defined as ‘the length of an equivalent pin-ended
column having the same load-carrying capacity as the member under consideration
provided with its actual conditions of support’. This engineering definition of effective length is illustrated in Fig. 15.7, which compares a column strength curve for a
member with some degree of rotational end restraint with the basic curve for the
same member when pin-ended.
In determining the column slenderness ratio the geometrical length, L, is replaced
by the effective length, LE. Values of effective length factors k = LE /L for a series
of standard cases are provided in BS 5950: Part 1, BS 5400: Part 3, etc.; Fig. 15.8 illustrates typical values. When compared with values given by elastic stability theory,2
these appear to be high for those cases in which reliance is being placed on externally provided rotational fixity; this is in recognition of the practical difficulties of
providing sufficient rotational restraint to approach the condition of full fixity. On
the other hand, translational restraints of comparatively modest stiffness are quite
capable of preventing lateral displacements. A certain degree of judgement is
required of the designer in deciding which of these standard cases most nearly
matches his arrangements. In cases of doubt the safe approach is to use a high
approximation, leading to an overestimate of column slenderness and thus an
underestimate of strength. The idea of an effective column length may also be used
as a device to deal with special types of column, such as compound or tapered
members, the idea then being to convert the complex problem into one of an equivalent simple column for which the basic design approach of the relationship between
compressive strength and slenderness may be employed.
Steel Designers' Manual - 6th Edition (2003)
414
Columns and struts
Pc for actual L/r
and effective LE/r
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
effective LE/r
N
N
column curve for end
restrained member
/
column
actual L/r
L/r
Fig. 15.7
Use of effective length with column curve to allow for end restraint
model
example
factor
1.0
0.85
I"
III
2.0
0.7
I; rl:rwi
Fig. 15.8
Typical effective length factors for use in strut design
1.0
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Effective lengths
415
Of fundamental importance when determining suitable effective lengths is the
classification of a column as either a sway case for which translation of one end
relative to the other is possible or a non-sway case for which end translation is prevented. For the first case, effective lengths will be at least equal to the geometrical
length, tending in theory to infinity for a pin-base column with no restraint at its
top, while for the non-sway case, effective lengths will not exceed the geometrical
length, decreasing as the degree of rotational fixity increases.
For non-standard cases, it is customary to make reference to published results
obtained from elastic stability theory. Provided these relate to cases for which buckling involves the interaction of a group of members with the less critical restraining
the more critical, as illustrated in Fig. 15.9, such evidence as is available suggests
that the use of effective lengths derived directly from elastic stability theory in conjunction with a column design curve of the type shown in Fig. 15.3 will lead to good
approximations of the true load-carrying capacity.
For compression members in rigid-jointed frames the effective length may, in both
cases, be directly related to the restraint provided by the surrounding members
by using charts presented in terms of the stiffness of these members provided in
Appendix E of BS 5950: Part 1. Useful guidance on effective column lengths for a
variety of more complex situations is available from several sources.2–4
When designing compression members in frames configured on the basis of
simple construction, the use of effective column lengths provides a simple means
of recognizing that real connections between members will normally provide some
degree of rotational end restraint, leading to compressive strengths somewhat in
excess of those that would be obtained if columns were treated as pin-ended. If axial
load levels and unsupported lengths change within the length of a member that is
continuous over several segments, such as a building frame column spliced so as to
act as a continuous member but carrying decreasing compression with height or a
compression chord in a truss, then the less heavily loaded segments will effectively
restrain the more critical segments.
Even though the distribution of internal member forces has been made on the
assumption of pin joints, some allowance for rotational end restraint when designing the compression members is therefore appropriate. Thus the apparent contradiction of regarding a structure as pin-jointed but using compression member
effective lengths that are less than their actual lengths does have a basis founded
upon an approximate version of reality. Figure 15.10 resents results obtained from
elastic stability theory for columns continuous over a number of storeys which show
LE
P
P
/
Fig. 15.9
L1
/
L
/
L1
Restraint to critical column segment from adjacent segments
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
416
Columns and struts
-2.
---..2_
N
-)
,-0
C.
—(0
.
.
—s.
E
0
!
T i T;
'
z
0((N(.4(0;;:;(.4N0).--.4-
0-0—..-—
•(0
(C)
•—
,-
0
I—I
-d.--
—
C
—
(C)
Cl)
-r)0
(N
(0
c.4
(-4J
GD
N
.N
•((N
)
(4
N((N
N
.—
(N
NC)
(N
(N
C)
NGD
GD
GD
C)
0)
(NNNNJ
N-
-dd
o
0
1.0
0.9 0.8 0.7
L
a
L
°
LI
.-
LI
I;
I
I
0.6 0.5 0.4 0.3 0.2 0.1
0
frame
column
Effective lengths
a/L
><
,,/\\
.j
><
L
°
Effective length factors for continuous columns based on elastic stability theory
1.0 0.97 0.94 0.91 0.88 0.85 0.82 0.79 0.76 0.73 0.70
L/L
Fig. 15.10
1.0 0.93 0.87 0.81 0.75 0.70 0.65 0.61 0.57 0.53 0.50
L/L
)(
V
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Steel Designers' Manual - 6th Edition (2003)
417
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
418
Columns and struts
how the effective length of the critical segment will be reduced if more stable segments (shorter unbraced lengths in this case) are present. A practical equivalent for
each case in terms of simple braced frames with pinned beam-to-column connections is also shown. It is also necessary to recognize that practical equivalents of pin
joints may also be capable of transferring limited moments. This point is considered
explicitly in BS 5950: Part 1 for both building frames and trusses; the effect on the
design of compression members is considered in detail in Chapter 18 for the former
and in Chapter 19 for the latter.
For compression members in rigid-jointed frames the effective length is directly
related to the restraint provided by all the surrounding members. Strictly speaking
an interaction of all the members in the frame occurs because the real behaviour is
one of frame buckling rather than column buckling, but for design purposes it is
often sufficient to consider the behaviour of a limited region of the frame. Variants
of the ‘limited frame’ concept are to be found in several codes of practice and
design guides. That used in BS 5950: Part 1 is illustrated in Fig. 15.11. The limited
frame comprises the column under consideration and each immediately adjacent
member treated as if its far end were fixed.The effective length of the critical column
is then obtained from a chart which is entered with two coefficients k1 and k2, the
values of which depend on the stiffnesses of the surrounding members KU, KTL, etc.,
relative to the stiffness of the column KC, a concept similar to the well-known
moment distribution method. Two distinct cases are considered: columns in nonsway frames and columns in frames that are free to sway. Figures 15.12 (a) and (b)
and 15.12 (c) and (d) illustrate both cases as well as giving the associated effective
2' '
(a)
Fig. 15.11
(b)
Limited frames. (a) Internal column. (b) External column
Steel Designers' Manual - 6th Edition (2003)
Effective lengths
pinned
KIL
/
-N
——
-— —.—
:;
0.8 -
KTR
K:
1 .C
, e\
' Ku
0.7
.
—ç
- -
k1
'..
H
fixed
0
--
N,;.\\
\
\\
.:
0.1
KBL
419
0
0.2
0.6
0.4
fixed
0.8 1.0
pinned
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(a)
(b)
pinned
-L
10
0.8
k1
07
Jr
0.,
:
:
0.2
o
....
.....
..
; z ::
N
..
.'
Fig. 15.12
\\\
\\
\
\ -. -\ -.
,
;., : \ \ \
\__'
.-
0
0.2
0.4
fixed
(c)
S
\
0.1
fixed
-
0
0.4
z
:
0.6
0.8 1.0
pinned
(d)
Limited frames and corresponding effective length charts of BS 5950: Part 1. (a)
Limited frame and (b) effective length ratios (k3 = •), for non-sway frames. (c)
Limited frame and (d) effective length ratios (without partial bracing, k3 = 0), for
sway frames
length charts. For the former, the factors will vary between 0.5 and 1.0 depending
on the values of k1 and k2, while for the latter, the variation will be between 1.0 and
•. These end points correspond to cases of: rotationally fixed ends with no sway and
rotationally free ends with no sway; rotationally fixed ends with free sway and rotationally free ends with free sway.
For beams not rigidly connected to the column or for situations in which significant plasticity either at a beam end or at either column end would prevent the
Steel Designers' Manual - 6th Edition (2003)
420
Columns and struts
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
restraint being transferred into the column, the K (and thus the k) values must be
suitably modified. Similarly at column bases, k2 values, in keeping with the degree
of restraint provided, should be used. Guidance is also provided on K values for
beams, distinguishing between both non-sway and sway cases and beams supporting concrete floors and bare steelwork. A further pair of charts permits modest
degrees of partial restraint against sway, as might be provided for example by infill
panels, to be allowed for in slightly reducing effective length values for sway frames.
Full details of the background to this approach to the determination of effective
lengths in rigidly jointed sway, partially braced and non-sway frames may be found
in the work of Wood.4
15.8 Special types of strut
The design of two types of strut requires that certain additional points be
considered:
(1) built-up sections or compound struts (Fig. 15.13), for which the behaviour of the
individual components must be taken into account
(2) angles, channels and tees (Fig. 15.14), for which the eccentricity of loading produced by normal forms of end connection must be acknowledged.
side column
Fig. 15.13
Built-up struts
valley column
Steel Designers' Manual - 6th Edition (2003)
Special types of strut
421
V
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Y
Fig. 15.14
'V
axis X—X
axis Y—Y
rectangular axes
axis U—U
axis V—V
maximum principal axis
minimum principal axis
Geometrical properties of an angle section
In both cases, however, it will often be possible to design this more complex type
of member as an equivalent single axially-loaded strut.
15.8.1 Design of compound struts
Individual members may be combined in a variety of ways to produce a more efficient compound section. Figure 15.15 illustrates the most common arrangements. In
each case the concept is one of providing a compound member whose overall slenderness will be such that its load-carrying capacity will significantly exceed the sum
of the axial resistances of the component members, i.e. for the case of Fig. 15.15(b)
the laced strut will be stronger than the four corner angles treated separately.
Thus the design approach of BS 5950: Part 1 is to set conditions which when met
permit the compound member to be designed as a single integral member. The following cases are considered explicitly:
(1) Laced struts conforming with the provisions of clause 4.7.8
(2) Battened struts conforming with the provisions of clause 4.7.9
(3) Batten-starred angles conforming with the provisions of clause 4.7.11, which
uses much of clause 4.7.9
(4) Batten parallel angle struts conforming with the provisions of clause 4.7.12
(5) Back-to-back struts conforming with the provisions of clause 4.7.13.1 if the components are separated and of clause 4.7.13.2 if the components are in contact.
The detailed rules contained in these clauses are essentially of two types:
Steel Designers' Manual - 6th Edition (2003)
422
Columns and struts
z
Y_JjL_Y
z
z
z
liL
JiL 1I
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
1i
z
z
(a)
y
y
X_[1f_X
y
y
y
-x
(b)
Fig. 15.15
Typical arrangements for compound struts: (a) closely spaced, (b) laced or
battened
(1) Covering construction details such as the arrangements for interconnection in
a general ‘good practice’ manner
(2) Quantitative rules for the determination of the overall slenderness, limits necessary for component slenderness, forces for which the interconnections should
be designed, etc.
Steel Designers' Manual - 6th Edition (2003)
Economic points
423
BS 5400: Part 3 also contains specific rules for the design of:
(1)
(2)
(3)
(4)
Batten struts (clause 10.8)
Laced struts (clause 10.9)
Struts connected by perforated cover plates (clause 10.10)
Struts consisting of back-to-back components (clause 10.11).
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
These are somewhat more detailed than those of BS 5950: Part 1, particularly in the
matter of determining suitable design forces for the interconnections, i.e. battens,
lacings, etc.
15.8.2 Design of angles, channels and tees
Four specific cases are covered in detail by BS 5950: Part 1:
(1)
(2)
(3)
(4)
Single angles in clause 4.7.10.2
Double angles in clause 4.7.10.3
Single channels in clause 4.7.10.4
Single tees in clause 4.7.10.5.
In all cases, guidance is provided on the determination of the slenderness to be used
when obtaining pc for each of the more common forms of fastening arrangement.
In cases where only a single fastener is used at each end the resulting value of pc
should then be reduced to 80% so as to allow for the combined effects of load eccentricity and lack of rotational end restraint. For ease of use the whole set of
slenderness relationships is grouped together in Table 25.
BS 5400: Part 3 only gives specific consideration to single angles.This distinguishes
between single bolt and ‘other’ forms of end connection. In both cases load eccentricity may be ignored but for the former only 80% of the calculated resistance may
be used.
15.9 Economic points
Strut design is a relatively straightforward design task involving choice of crosssectional type, assessment of end restraint and thus effective length, calculation of
slenderness, determination of compressive strength and hence checking that the trial
section can withstand the design load. Certain subsidiary checks may also be
required part way through this process to ensure that the chosen cross section is not
slender (or make suitable allowances if it is) or to guard against local failure in compound members.Thus only limited opportunities occur for the designer to use judgement and to make choices on the grounds of economy. Essentially these are
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
424
Columns and struts
restricted to control of the effective length, by introducing intermediate restraints
where appropriate, and the original choice of cross section.
However, certain other points relating to columns may well have a bearing on the
overall economy of the steel frame or truss. Of particular concern is the need to be
able to make connections simply. In a multi-storey frame, the use of heavier UC
sections thus may be advantageous, permitting beam-to-column connections to be
made without the need for stiffening the flanges or web. Similarly in order to
accommodate beams framing into the column web an increase in the size of UC
may eliminate the need for special detailing.
While compound angle members were a common feature of early trusses, maintenance costs due both to the surface area requiring painting and to the incidence
of corrosion caused by the inherent dirt and moisture traps have caused a change
to the much greater use of tubular members. If site joints are kept to the minimum
tubular trusses can be transported and handled on site in long lengths and a more
economic as well as a visually more pleasing structure is likely to result.
References to Chapter 15
1. The Steel Construction Institute (SCI) (2001) Steelwork Design Guide to BS
5950: Part 1: 2000 Vol. 1: Section Properties, Member Capacities, 6th edn. SCI,
Ascot, Berks.
2. Ballio G. & Mazzolani F.M. (1983) Theory and Design of Steel Structures.
Chapman and Hall.
3. Allen H.G. & Bulson P.S. (1980) Background to Buckling. McGraw-Hill.
4. Wood R.H. (1974) Effective lengths of columns in multi-storey buildings. The
Structural Engineer, 52, Part 1, July, 235–44, Part 2, Aug., 295–302, Part 3, Sept.,
341–6.
Further reading for Chapter 15
ECCS (1986) Behaviour and Design of Steel Plated Structures (Ed. by P. Dubas &
E. Gehri). ECCS Publication No. 44.
Galambos T.V. (Ed.) (1998) Guide to Stability Design Criteria for Metal Structures,
5th edn. Wiley, New York.
Hancock G.J. (1988) Design of Cold-Formed Steel Structures. Australian Institute of
Steel Construction, Sydney.
Kirby P.A. & Nethercot D.A. (1985) Design for Structural Stability. Collins, London.
Trahair N.S. & Nethercot D.A. (1984) Bracing Requirements in Thin-Walled
Structures, Developments in Thin-Walled Structures – 2 (Ed. by J. Rhodes & A.C.
Walker), pp. 92–130. Elsevier Applied Science Publishers, Barking, Essex.
A series of worked examples follows which are relevant to Chapter 15.
Steel Designers' Manual - 6th Edition (2003)
Worked examples
Subject
The
Steel Construction
Institute
Chapter ref.
COLUMN EXAMPLE 1
ROLLED UNIVERSAL
COLUMN
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
425
Made by DAN
15
Sheet no.
1
Checked by GWO
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Problem
Check the ability of a 203 ¥ 203 ¥ 52 UC in S275 steel to withstand
an axial compressive load of 1250 kN over an unsupported height of
3.6 m assuming that both ends are held in position but are provided
with no restraint in direction. Design to BS 5950: Part 1.
The problem is as shown in the sketch.
3.6m
II
Take effective length
\
LE = 1.0 L
LE
Table 22
= 1.0 ¥ 3.6 ¥ 103
= 3600 mm
On the assumption that weak axis flexural buckling will govern
determine compressive strength pc from Table 24 curve c
A = 66.4 cm2
Table 23
b/T
= 8.16
d/t
= 20.1
py
= 275 N/mm2
Table 9
Check section classification for pure compression.
3.5
ry
= 5.16 cm
Since T < 16 mm
take
Need only check section is not slender;
for outstand
b/T £ 15e
Steelwork
Design Guide
Vol 1
Table 11
Steel Designers' Manual - 6th Edition (2003)
Worked examples
426
Subject
The
Steel Construction
Institute
Chapter ref.
COLUMN EXAMPLE 1
ROLLED UNIVERSAL
COLUMN
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
for web
e
d/t
Sheet no.
2
Checked by GWO
£ 40e
( 275 py )
=
Made by DAN
15
=
1.0
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
actual b/T = 8.16, within limit
= 20.1, within limit
actual d/t
\ Section is not slender.
Pc = Agpc
l
4.7.4
= LE/ry
= 3600/51.6
For
l
= 70
value of
pc
= 202 N/mm2
\
and
Pc = 6640 ¥ 202
= 70
py
= 275 N/mm2
= 1341 ¥ 103 N
= 1341 kN
This exceeds required resistance of 1250 kN and section is therefore
OK.
\ Use 203 ¥ 203 ¥ 52 UC
4.7.3
Table 24
Steel Designers' Manual - 6th Edition (2003)
Worked examples
Subject
The
Steel Construction
Institute
Chapter ref.
COLUMN EXAMPLE 2
WELDED BOX
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
427
Made by DAN
15
Sheet no.
1
Checked by GWO
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Problem
Check the ability of a 960 mm square box column fabricated from
30 mm thick S355 plate to withstand an axial compressive load of
22000 kN over an unsupported height of 15 m assuming that both
ends are held in position but are provided with no restraint in direction. Design to BS 5950: Part 1.
The problem is as shown in the sketches.
960 mi
mI
LE = 1.0 L
Take effective length
\
LE
Table 22
= 1.0 ¥ 15 ¥ 103
= 15000 mm
Determine pc from Table 24, curve b
Since T > 16 mm, take py
= 345 N/mm2
Check section classification for pure compression.
Need only check section is not slender;
for flange
e
=
b/T £ 40e
( 275 345 )
=
0.9
Table 23
Table 11
3.5
Table 11
Steel Designers' Manual - 6th Edition (2003)
Worked examples
428
Subject
The
Steel Construction
Institute
COLUMN EXAMPLE 2
WELDED BOX
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
actual b/T = (960 - 2 ¥ 30)/30
= 0.9 ¥ 40
limit
Chapter ref.
Made by DAN
15
Sheet no.
2
Checked by GWO
= 30
= 36.0
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
\ Section is not slender.
Iy
=
( 960 ¥ 960 3 - 900 ¥ 900 3 ) 12 = 1.61 ¥ 1010 mm 4
Ag
=
2 ¥ 30( 960 + 900 )
ry
=
l
=
=
111600 mm2
(1.61 ¥ 1010 111600 ) = 380 mm
LE ry
= 15000 380
= 39.5
4.3.7.1
Since section fabricated by welding, use Table 24 curve b with
py
4.7.5
= 345 - 20
= 325 N/mm2
For l = 39.5
value of
and
py
= 325 N/mm2
Table 27b
pc = 293 N/mm2
Pc = Agpc = 111600 ¥ 293
= 32.7 ¥ 106 N
= 32700 kN
This exceeds required resistance of 22000 kN and section is OK.
\ Use section shown in figure.
4.7.4
Steel Designers' Manual - 6th Edition (2003)
Worked examples
Subject
The
Steel Construction
Institute
Chapter ref.
COLUMN EXAMPLE 3
ROLLED UNIVERSAL BEAM
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
15
Made by DAN
Sheet no.
1
Checked by GWO
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Problem
A 457 ¥ 191 ¥ 89 UB in S275 steel is to be used as an axially loaded
column over a free height of 7.0 m. Both ends will be held in position but not direction for both planes. The possibility exists to
provide discrete bracing members capable of preventing deflection in
the plane of the flanges only. Investigate the advisability of using
such bracing.
Clearly strength cannot exceed
that for major axis failure.
Check no. of intermediate
(minor-axis) braces needed to
achieve this.
Take
\
LEx
I
major axis
LEx = 1.0 ¥ 7.0 ¥ 103 = 7000 mm
A = 114 cm2
ry
minor axis
= 1.0 L
= 4.28 cm
rx = 19.0 cm
Section not slender
Pc = Agpc
lx = LEx/rx
b/T
= 5.42
d/t
= 38.5
py
= 265 N/mm2 (for T = 17.7 mm)
Steelwork
Design Guide
Vol 1
Table 11
4.7.4
= 7000/190
= 37
4.7.3
Use Table 24, curve a for pcx
Table 23
429
Steel Designers' Manual - 6th Edition (2003)
430
Worked examples
Subject
The
Steel Construction
Institute
COLUMN EXAMPLE 3
ROLLED UNIVERSAL BEAM
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
For
l = 37
Chapter ref.
and
Made by DAN
Sheet no.
2
Checked by GWO
= 265 N/mm2
py
15
Table 24
pcx = 253 N/mm2
value of
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Pcx = 11400 ¥ 253 = 2.884 ¥ 106 N
= 2884 kN
Use Table 24, curve b for pcy
& Table 23
For a pcy value of 253 N/mm2, value of ly 30
\
LEy /42.8 = 30
and
LEy 1284 mm
Table 24
\ to achieve a minor axis buckling resistance 2987 kN would
require 7.0 m height to be provided with 7000/1284 Æ 5 restraints
LEy /ry
= (7000/6)/42.8 = 27
pcy
= 256 N/mm2
Pcy
= 11400 ¥ 256
Table 24
= 2.918 ¥ 106
= 2918 kN
Not however that substantial improvements on the basic minor axis
resistance for LEy = 7.0 m of 741 kN may be achieved for rather less
restraints as shown below.
no. of restraints
0
1
2
3
Pcy (kN)
Pcy /Pcx
730
1972
2531
2645
0.25
0.68
0.88
0.92
Steel Designers' Manual - 6th Edition (2003)
Chapter 16
Beams
by DAVID NETHERCOT
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
16.1 Common types of beam
Beams are possibly the most fundamental type of member present in a civil engineering structure. Their principal function is the transmission of vertical load by
means of flexural (bending) action into, for example, the columns in a rectangular
building frame or the abutments in a bridge which support them.
Table 16.1 provides some idea of the different structural forms suitable for use
as beams in a steel structure; several of these are illustrated in Fig. 16.1. For modest
spans, including the majority of those found in buildings, the use of standard
hot-rolled sections (normally UBs but possibly UCs if minimizing floor depth is a
prime consideration or channels if only light loads need to be supported) will
be sufficient. Lightly loaded members such as the purlins supporting the roof
of a portal-frame building are frequently selected from the range of proprietary
cold-formed sections produced from steel sheet only a few millimetres thick, normally already protected against corrosion by galvanizing, in a variety of highly efficient shapes, advantage being taken of the roll-forming process to produce sections
with properties carefully selected for the task they are required to perform. For
spans in excess of those that can be achieved sensibly using ready-made sections
some form of built-up member is required. Castellated beams, formed by profile
cutting of the web and welding to produce a deeper section, typically 50% deeper
using the standard UK geometry, are visually attractive but cannot withstand high
shear loads unless certain of the castellations are filled in with plate. The range of
spans for which UBs may be used can be extended if cover plates are welded to
both flanges.
Alternatively a beam fabricated entirely by welding plates together may be
employed allowing variations in properties by changes in depth, for example, flange
thickness, or, in certain cases where the use of very thin webs is required, stiffening
to prevent premature buckling failure is necessary. A full treatment of the specialist aspects of plate-girder design is provided in Chapter 17. If spans are so large that
a single member cannot economically be provided, then a truss may be a suitable
alternative. In addition to the deep truss fabricated from open hot-rolled sections,
SHS or both, used to provide long clear spans in sports halls and supermarkets,
smaller prefabricated arrangements using RHS or CHS provide an attractive alternative to the use of standard sections for more modest spans. Truss design is discussed in Chapter 19.
431
Steel Designers' Manual - 6th Edition (2003)
432
Beams
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Table 16.1 Typical usage of different forms of beam
Beam type
Span range (m)
Notes
(0) Angles
3–6
(1) Cold-formed
sections
(2) Rolled
sections:
UBs, UCs,
RSJs, RSCs
(3) Open web
joists
(4) Castellated
beams
4–8
Used for roof purlins, sheeting rails, etc., where only light
loads have to be carried
Used for roof purlins, sheeting rails, etc., where only light
loads have to be carried
Most frequently used type of section; proportions selected to
eliminate several possible types of failure
1–30
4–40
6–60
(5) Compound
sections e.g.
UB + RSC
(6) Plate girders
5–15
10–100
(7) Trusses
10–100
(8) Box girders
15–200
Prefabricated using angles or tubes as chords and round bar
for web diagonals, used in place of rolled sections
Used for long spans and/or light loads; depth of UB
increased by 50%; web openings may be used for
services, etc.
Used when a single rolled section would not provide
sufficient capacity; often arranged to provide enhanced
horizontal bending strength as well
Made by welding together 3 plates sometimes automatically;
web depths up to 3–4 m sometimes need stiffening
Heavier version of (3); may be made from tubes, angles or,
if spanning large distances, rolled sections
Fabricated from plate, usually stiffened; used for OHT cranes
and bridges due to good torsional and transverse stiffness
properties
Since the principal requirement of a beam is adequate resistance to vertical
bending, a very useful indication of the size of section likely to prove suitable may
be obtained through the concept of the span to depth ratio. This is simply the value
of the clear span divided by the overall depth. An average figure for a properly
designed steel beam is between 15 and 20, perhaps more if a particularly slender
form of construction is employed or possibly less if very heavy loadings are present.
When designing beams, attention must be given to a series of issues, in addition
to simple vertical bending, that may have some bearing upon the problem. Torsional
loading may often be eliminated by careful detailing or its effects reasonably
regarded as of negligible importance by a correct appreciation of how the structure
actually behaves; in certain instances it should, however, be considered. Section classification (allowance for possible local buckling effects) is more involved than is the
case for struts since different elements in the cross-section are subject to different
patterns of stress; the flanges in an I-beam in the elastic range will be in approximately uniform tension or compression while the web will contain a stress gradient.
The possibility of members being designed for an elastic or a plastic state, including the use of a full plastic design for the complete structure, also affects section
classification. Various forms of instability of the beam as a whole or of parts subject
to locally high stresses, such as the web over a support, also require attention. Finally
certain forms of construction may lend themselves to the appearance of unacceptable vibrations; although this is likely to be affected by the choice of beams, its coverage is left to Chapter 20 on floors.
Steel Designers' Manual - 6th Edition (2003)
Cross-section classification and moment capacity, Mc
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
rolled steel channel
universal beam
plate girder
433
cold— formed section
castellated section
box girder
Fig. 16.1
Types of beam cross-section
16.2 Cross-section classification and moment capacity, Mc
The possible influences of local buckling on the ability of a particular cross-section
to attain, and where appropriate also to maintain, a certain level of moment are discussed in general terms in Chapter 13. In particular, section 13.2 covers the influence of flange (b/T) and web (d/t) slenderness on moment – rotation behaviour (Fig.
13.4) and moment capacity (Fig. 13.5). When designing beams it is usually sufficient
to consider the web and the compression flange separately using the appropriate
sets of limits.1
Steel Designers' Manual - 6th Edition (2003)
434
Beams
In building design when using hot-rolled sections, for which relevant properties
are tabulated,2 it will normally be sufficient to ascertain a section’s classification and
moment capacity simply by referring to the appropriate table. An example illustrates the point.
Example
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Using BS 5950: Part 11 determine the section classification and moment capacity for
a 533 ¥ 210 UB82 when used as a beam in (1) S275 steel and (2) S355 steel.
(1) From Reference 2, p. 129, section is ‘plastic’ and Mcx = 566 kN m
Alternatively, from Reference 2, p. 26, b/T = 7.91, d/t = 49.6
1
Since T = 13.2 mm from Reference 1, Table 9, py = 275 N/mm2 and e = (275/py)–2
= 1.0
From Reference 1, Table 11, limits for plastic section are b/t 9e, d/t 80e;
actual b/t and d/t are within these limits and section is plastic.
From Reference 1, clause 4.2.5, Mc = py S
= 275 ¥ 2060 ¥ 10-3
= 566.5 kNm
(2) From Reference 2, p. 303, section is compact and Mcx = 731 kN m
Alternatively, from Reference 2, p. 26, b/T = 7.91, d/t = 49.6
1
Since T = 13.2 mm from Reference 1, Table 9, py = 355 N/mm2 and e = (275/py)–2
= 0.88
From Reference 1, Table 11, limits for plastic section are b/T 9e, d/t = 80e,
which give b/T 7.9, d/t 72; actual b/T of 7.91 exceeds 7.9 so check compact
limits of b/T 10e, which gives b/T 9; actual b/T of 7.91 meets this and section
is compact
From Reference 1, clause 4.2.5, Mc = py S
= 355 ¥ 2060 ¥ 10-3
= 731.3 kN m
Thus whichever grade of steel is being used, the moment capacity, Mc, will be the
maximum attainable value corresponding to the section’s full plastic moment capacity, Mp. Provided that S275 material is used, the beam is capable of redistributing
moments, since the plastic cross section behaves as illustrated in Fig. 13.4(a) and
so could be used in a plastically designed frame. Use of the higher strength S355
material, while not affecting the beam’s ability to attain Mp, precludes the use of
plastic design since redistribution of moments cannot occur due to the lack of rotation capacity implied by the behaviour illustrated in Fig. 13.4(b). For a continuous
structure designed on the basis of elastic analysis to determine the distribution of
internal forces and moments or for simple construction, the question of rotation
capacity is not relevant and in both cases moment capacity should be taken as Mp.
The fact that the section is plastic for S275 material is then of no particular relevance;
a more appropriate classification would be to regard the section as compact or better.
Steel Designers' Manual - 6th Edition (2003)
Cross-section classification and moment capacity, Mc
435
Table 16.2 Non-plastic UBs and UCs in bending
Compact
Semi-compact
S275
UC
305 ¥ 305 ¥ 97
203 ¥ 203 ¥ 46
356 ¥ 368 ¥ 129
152 ¥ 152 ¥ 23
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
S355
UB
762
406
406
305
254
203
UC
305 ¥ 305 ¥ 118
203 ¥ 203 ¥ 52
152 ¥ 152 ¥ 30
¥
¥
¥
¥
¥
¥
267
178
140
165
146
133
¥
¥
¥
¥
¥
¥
134
54
39
40
31
25
356 ¥ 171 ¥ 45
356
356
305
254
203
152
¥
¥
¥
¥
¥
¥
368
368
305
254
203
152
¥
¥
¥
¥
¥
¥
153
129
97
73
46
23
Inspection of the relevant tables in Reference 2 shows that when used as beams:
S275 steel
all UBs are plastic
all but 4 UCs are plastic
(2 are semi-compact)
S355 steel
all but 7 UBs are plastic
(1 is semi-compact)
all but 9 UCs are plastic
(6 are semi-compact)
In those cases of semi-compact sections, moment capacities based on My are listed.
The full list of non-plastic beam sections is given as Table 16.2.
When using fabricated sections individual checks on the web and compression
flange using the actual dimensions of the trial section must be made. It normally
proves much simpler if proportions are selected so as to ensure that the section is
compact or better since the resulting calculations need not then involve the various
complications associated with the use of non-compact sections. This point is particularly noticeable when designing bridge beams to BS 5400: Part 3. When slender
sections are used the amount of calculation increases considerably due to the need
to consider loss of effectiveness of some parts of the cross section due to local buckling when determining Mc. Probably the most frequent use of slender sections
involves cold-formed shapes used for example as roof purlins. Because of their
Steel Designers' Manual - 6th Edition (2003)
436
Beams
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
proprietary nature, the manufacturers normally provide design information, much
of it based on physical testing, listing such properties as moment capacity. In the
absence of design information, reference should be made to Part 5 of BS 5950 for
suitable calculation methods.
Although the part of the web between the flange and the horizontal edge of the
castellation in a castellated beam frequently exceeds the compact limit for an outstand, sufficient test data exist to show that this does not appear to influence the
moment capacity of such sections. Section classification should therefore be made
in the same way as for solid web beams, the value of Mc being obtained using the
net modulus value for the section at the centre of a castellation.
16.3 Basic design
One (or more) of a number of distinct limiting conditions may, in theory, control
the design of a particular beam as indicated in Table 16.3, but in any particular practical case only a few of them are likely to require full checks. It is therefore convenient to consider the various possibilities in turn, noting the conditions under
which each is likely to be important. For convenience the various phenomena are
first considered principally within the context of using standard hot-rolled sections,
i.e. UBs, UCs, RSJs and channels; other types of cross-section are covered in the
later parts of this chapter.
Table 16.3 Limiting conditions for beam design
Ultimate
Moment capacity, Me (including
influence of local buckling)
Shear capacity, Pv
Lateral – torsional buckling, Mb
Web buckling, Pw
Web bearing, Pyw
Moment – shear interaction
Torsional capacity, MT
Bending – torsion interaction
Serviceability
Deflections due to bending
(and shear if appropriate)
Twist due to torsion
Vibration
16.3.1 Moment capacity, Mc
The most basic design requirement for a beam is the provision of adequate in-plane
bending strength. This is provided by ensuring that Mc for the selected section
exceeds the maximum moment produced by the factored loading. Determination of
Mc, which is linked to section classification, is fully covered in section 16.2.
Steel Designers' Manual - 6th Edition (2003)
Basic design
437
For a statically determinate structural arrangement, simple considerations of
statics provide the moment levels produced by the applied loads against which Mc
must be checked. For indeterminate arrangements, a suitable method of elastic
analysis such as moment distribution or slope deflection is required. The justification for using an elastically obtained distribution of moments with, in the case of
compact or plastic cross-sections, a plastic cross-sectional resistance has been fully
discussed by Johnson and Buckby.3
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
16.3.2 Effect of shear
Only in cases of high coincident shear and moment, found for example at the internal supports of continuous beams, is the effect of shear likely to have a significant
influence on the design of beams.
Shear capacity Pv is normally calculated as the product of a shear strength pv,
often taken for convenience as 0.6py which is close to the yield stress of steel in
shear of 1/÷3 of the uniaxial tensile yield stress, and an appropriate shear area Av.
The process approximates the actual distribution of shear stress in a beam web as
well as assuming some degree of plasticity. While suitable for rolled sections, it may
not therefore be applicable to plate girders. An alternative design approach, more
suited to webs containing large holes or having variations in thickness, is to work
from first principles and to limit the maximum shear stress to a suitable value; BS
5950: Part 1 uses 0.7py. In cases where d/t > 63, shear buckling limits the effectiveness of the web and reference to Chapter 17 should be made for methods of determining the reduced capacity.
In principle the presence of shear in a section reduces its moment capacity. In
practice the reduction may be regarded as negligibly small up to quite large fractions of the shear capacity Pv. For example, BS 5950: Part 1 requires a reduction in
Mc for plastic or compact sections only when the applied shear exceeds 0.6Pv, and
permits the full value of Mc to be used for all cases of semi-compact or slender
sections.
Figure 16.2 illustrates the application of the BS 5950: Part 1 rule for plastic or
compact sections to a typical UB and UC having approximately equal values of
plastic section modulus Sx. Evaluation of the formula in cases where Fv/Pv > 0.6 first
requires that the value of Sv, the plastic modulus of the shear area Av (equal to tD
in this case), be determined. This is readily obtained from the tabulated values of S
for rectangles given in Reference 2 corresponding to a linear reduction from Sx to
(Sx - Sv).
16.3.3 Deflection
When designing according to limit state principles it is customary to check that
deflections at working load levels will not be such as to impair the proper function
Steel Designers' Manual - 6th Edition (2003)
Beams
438
1.0
-0 (I), 0.8
0
shear
cc)
C
—
> Q_
',.
E2.
area A
0.6
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
.LD
plastic modulus for
::
whole section
plastic modulus for
shear area
0
0.2
0.4
0.6
0.8
1.0
proportion of shear capacity present
F /P
Fig. 16.2
Moment – shear interaction for plastic or compact sections to BS 5950: Part I
of the structure. For beams, examples of potentially undesirable consequences of
excessive serviceability deflections include:
•
•
•
cracking of plaster ceilings
allowing crane rails to become misaligned
causing difficulty in opening large doors.
Although earlier codes of practice specified limits for working load deflections, the
tendency with more recent documents1 is to draw attention to the need for deflection checks and to provide advisory limits to be used only when more specific guidance is not available. Table 8 of BS 5950: Part 1 gives ‘recommended limitations’ for
certain types of beams and crane gantry girders and states that ‘Circumstances may
arise where greater or lesser values would be more appropriate.’ Not surprisingly
surveys of current practice4 reveal large variations in what is considered appropriate
for different circumstances.
When checking deflections of steel structures under serviceability loading, the
central deflection Dmax of a uniformly-loaded simply-supported beam, assuming
linear–elastic behaviour, is given by
Steel Designers' Manual - 6th Edition (2003)
Basic design
439
Table 16.4 Minimum I values for uniformly-loaded simply-supported beams for various deflection limits
a
K
200
1.24
D max =
240
1.49
250
1.55
325
2.02
360
2.23
5 WL3 12
10 (mm)
384 EI
400
2.48
500
3.10
600
3.72
750
4.65
1000
6.20
(16.1)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
in which W = total load (kN)
E = Young’s modulus (N/mm2)
I = second moment of area (mm4)
L = span (m).
If Dmax is to be limited to a fraction of L, Equation (16.1) may be rearranged to
give
Irqd = 0.62 ¥ 10-2 aWL2
(16.2)
in which a defines the deflection limit as
D max = L /a
and Irqd is now in cm4.
Writing Irqd = KWL2, Table 16.4 gives values of K for a range of values of a.
Since deflection checks are essentially of the ‘not greater than’ type, some degree
of approximation normally is acceptable, particularly if the calculations are reduced
as a result. Converting complex load arrangements to a roughly equivalent UDL
permits Table 16.4 to be used for a wide range of practical situations. Table 16.5 gives
values of the coefficient K by which the actual load arrangement shown should be
multiplied in order to obtain an approximately equal maximum deflection.
Tables of deflections for a number of standard cases are provided in the
Appendix.
16.3.4 Torsion
Beams subjected to loads which do not act through the point on the cross-section
known as the shear centre normally suffer some twisting. Methods for locating the
shear centre for a variety of sectional shapes are given in Reference 5. For doubly
symmetrical sections such as UBs and UCs it coincides with the centroid while for
channels it is situated on the opposite side of the web from the centroid; for rolled
channels its location is included in the tables of Reference 2. Figure 16.3 illustrates
its position for a number of standard cases.
The effects of torsional loading may often be minimized by careful detailing, particularly when considering how loads are transferred between members. Proper
attention to detail can frequently lead to arrangements in which the load transfer
Steel Designers' Manual - 6th Edition (2003)
Beams
440
Table 16.5 Equivalent UDL coefficients
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
0
A
'I
0
a
a/L
KD
0.5
0.4
0.375
0.333
0.3
0.25
0.2
0.1
1.0
0.86
0.82
0.74
0.68
0.58
0.47
0.24
A
A
b
1—0
C
C
C
C
C
b
C
I
No. of equal loads
b/L
c/L
K
2
0.2
0.25
0.333
0.167
0.2
0.25
0.125
0.2
0.1
0.167
0.083
0.143
0.071
0.125
0.063
0.111
0.6
0.5
0.333
0.333
0.3
0.25
0.25
0.2
0.2
0.167
0.167
0.143
0.143
0.125
0.125
0.111
0.91
1.10
1.3
1.05
1.14
1.27
1.03
1.21
1.02
1.17
1.01
1.15
1.01
1.12
1.01
1.11
3
4
5
6
7
8
I'
a
F-
K D = 1.6
a/L
KD
a
L
a
-1
2
È3 - 4 a ˘
ÍÎ
L ˙˚
0.01
0.05
()
0.05
0.24
0.1
0.47
0.15
0.70
0.2
0.91
0.25
1.10
0.3
1.27
0.35
1.41
0.4
1.51
0.45
1.58
0.5
1.60
is organized in such a way that twisting should not occur.5 Whenever possible this
approach should be followed as the open sections normally used as beams are inherently weak in resisting torsion. In circumstances where beams are required to withstand significant torsional loading, consideration should be given to the use of a
torsionally more efficient shape such as a structural hollow section.
16.3.5 Local effects on webs
At points within the length of a beam where vertical loads act, the web is subject to concentrations of stresses, additional to those produced by overall
bending. Failure by buckling, rather in the manner of a vertical strut, or by the development of unacceptably high bearing stresses in the relatively thin web material
Steel Designers' Manual - 6th Edition (2003)
Basic design
H=O
SI
e=
F+ 6
H= a2b3t2
12
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
b
aLl
\F+6
where F at1
—
bt2
H=1
e—
ly
02
H
j1 j2
ly
I and '2 are the respective
second moments of area of the
where
flanges about the YY axis
a2b3t2 (2o+b
H— 12 o+2b
3S
e=
o2b2t fl c
2c3
x
H= -(4c3+6oc2+3a2c+o2b)_e21x
Fig. 16.3
Location of shear centre for standard sections (H is warping constant)
441
Steel Designers' Manual - 6th Edition (2003)
442
Beams
e=
xtxo
cii
Jx
2c3
H= (4c3_6oc2+3a2c+a2b)_e2Ix
021y
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
\ b)
o2b2t (!+ C
+c2b2t(+)
/
b2t l+4c2(3bo÷3o2+4bc+2oc+c2)
a(b+2ba+4bc+6oc)
H=
1
Fig. 16.3
2(2b+a+2c)
(continued)
immediately adjacent to the flange, is a possibility. Methods for assessing the likelihood of both types of failure are given in BS 5950: Part 1, and tabulated data to
assist in the evaluation of the formulae required are provided for rolled sections in
Reference 2. The parallel approach for cold-formed sections is discussed in section
16.7.
In cases where the web is found to be incapable of resisting the required level of
load, additional strength may be provided through the use of stiffeners. The design
of load-carrying stiffeners (to resist web buckling) and bearing stiffeners is covered
in both BS 5950 and BS 5400. However, web stiffeners may be required to resist
shear buckling, to provide torsional support at bearings or for other reasons; a full
treatment of their design is provided in Chapter 17.
16.3.6 Lateral – torsional buckling
Beams for which none of the conditions listed in Table 16.6 are met (explanation
of these requirements is delayed until section 16.3.7 so that the basic ideas and
parameters governing lateral – torsional buckling may be presented first) are liable
to have their load-carrying capacity governed by the type of failure illustrated in
Fig. 16.4. Lateral – torsional instability is normally associated with beams subject to
Steel Designers' Manual - 6th Edition (2003)
Basic design
443
Table 16.6 Types of beam not susceptible to lateral – torsional buckling
loading produces bending about the minor axis
beam provided with closely spaced or continuous lateral restraint
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
x
C
N
closed section
Fig. 16.4
Lateral – torsional buckling
vertical loading buckling out of the plane of the applied loads by deflecting
sideways and twisting; behaviour analogous to the flexural buckling of struts. The
presence of both lateral and torsional deformations does cause both the governing
mathematics and the resulting design treatment to be rather more complex.
The design of a beam taking into account lateral – torsional buckling consists
essentially of assessing the maximum moment that can safely be carried from a
knowledge of the section’s material and geometrical properties, the support conditions provided and the arrangement of the applied loading. Codes of practice, such
as BS 5400: Part 3, BS 5950: Parts 1 and 5, include detailed guidance on the subject.
Essentially the basic steps required to check a trial section (using BS 5950: Part I
for a UB as an example) are:
(1) assess the beam’s effective length LE from a knowledge of the support conditions provided (clause 4.3.5)
(2) determine beam slenderness lLT using the geometrical parameters u (tabulated
in Reference 2), LE/ry, v (Table 19 of BS 5950: Part 1) using values of x (tabulated in Reference 2).
(3) obtain corresponding bending strength pb (Table 16)
(4) calculate buckling resistance moment Mb = pb ¥ the appropriate section
modulus, Sx (class 1 or 2), Zx (class 3), Zx,eff (class 4).
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
444
Beams
The central feature in the above process is the determination of a measure of the
beam’s lateral – torsional buckling strength (pb) in terms of a parameter (lLT) which
represents those factors which control this strength. Modifications to the basic
process permit the method to be used for unequal flanged sections including tees,
fabricated Is for which the section properties must be calculated, sections containing slender plate elements, members with properties that vary along their length,
closed sections and flats. Various techniques for allowing for the form of the applied
loading are also possible; some care is required in their use.
The relationship between pb and lLT of BS 5950: Part 1 (and between sli/syc and
lLT ÷(syc/355) in BS 5400: Part 3) assumes the beam between lateral restraints to be
subject to uniform moment. Other patterns, such as a linear moment gradient reducing from a maximum at one end or the parabolic distribution produced by a uniform
load, are generally less severe in terms of their effect on lateral stability; a given
beam is likely to be able to withstand a larger peak moment before becoming laterally unstable. One means of allowing for this in design is to adjust the beam’s slenderness by a factor n, the value of which has been selected so as to ensure that the
resulting value of pb correctly reflects the enhanced strength due to the non-uniform
moment loading. An alternative approach consists of basing lLT on the geometrical
and support conditions alone but making allowance for the beneficial effects of nonuniform moment by comparing the resulting value of Mb with a suitably adjusted
value of design moment M . M is taken as a factor m times the maximum moment
within the beam Mmax; m = 1.0 for uniform moment and m < 1.0 for non-uniform
moment. Provided that suitably chosen values of m and n are used, both methods
can be made to yield identical results; the difference arises simply in the way in
which the correction is made, whether on the slenderness axis of the pb versus lLT
relationship for the n-factor method or on the strength axis for the m-factor method.
Figure 16.5 illustrates both concepts, although for the purpose of the figure the mfactor method has been shown as an enhancement of pb by 1/m rather than a reduction in the requirement of checking Mb against M = mMmax. BS 5950: Part 1 uses the
m-factor method for all cases, while BS 5400: Part 3 includes only the n-factor
method.
When the m-factor method is used the buckling check is conducted in terms of
a moment M less than the maximum moment in the beam segment Mmax; then a
separate check that the capacity of the beam cross-section Mc is at least equal to
Mmax must also be made. In cases where M is taken as Mmax, then the buckling
check will be more severe than (or in the ease of a stocky beam for which Mb = Mc,
identical to) the cross-section capacity check.
Allowance for non-uniform moment loading on cantilevers is normally treated
somewhat differently. For example, the set of effective length factors given in Table
14 of Reference 1 includes allowances for the variation from the arrangement used
as the basis for the strength – slenderness relationship due to both the lateral support
conditions and the form of the applied loading. When a cantilever is subdivided by
one or more intermediate lateral restraints positioned between its root and tip, then
segments other than the tip segment should be treated as ordinary beam segments
when assessing lateral – torsional buckling strength. Similarly a cantilever subject to
Steel Designers' Manual - 6th Edition (2003)
Basic design
445
effective design
point using
n—factor method
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(Pb' nXLT)
LI
Fig. 16.5
Design modifications using m-factor or n-factor methods
an end moment such as horizontal wind load acting on a façade, should be regarded
as an ordinary beam since it does not have the benefit of non-uniform moment
loading.
For more complex arrangements that cannot reasonably be approximated by one
of the standard cases covered by correction factors, codes normally permit the direct
use of the elastic critical moment ME. Values of ME may conveniently be obtained
from summaries of research data.6 For example, BS 5950: Part 1 permits lLT to be
calculated from
l LT =÷(p 2 E / py ) ÷(M p / M E )
(16.3)
As an example of the use of this approach Fig. 16.6 shows how significantly higher
load-carrying capacities may be obtained for a cantilever with a tip load applied to
its bottom flange, a case not specifically covered by BS 5950: Part 1.
16.3.7 Fully restrained beams
The design of beams is considerably simplified if lateral – torsional buckling effects
do not have to be considered explicitly – a situation which will occur if one or more
of the conditions of Table 16.6 are met.
In these cases the beam’s buckling resistance moment Mb may be taken as its
moment capacity Mc and, in the absence of any reductions in Mc due to local buck-
Steel Designers' Manual - 6th Edition (2003)
446
Beams
UB 610 x 229 x 101
1.0
at bottom flange
0.8
load at centroid
0.6
0,4
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
0.2
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
2.6
XLI
Fig. 16.6
Lateral – torsional buckling of a tip-loaded cantilever
ling, high shear or torsion, it should be designed for its full in-plane bending strength.
Certain of the conditions corresponding to the case where a beam may be regarded
as ‘fully restrained’ are virtually self-evident but others require either judgement or
calculation.
Lateral – torsional buckling cannot occur in beams loaded in their weaker principal plane; under the action of increasing load they will collapse simply by plastic
action and excessive in-plane deformation. Much the same is true for rectangular
box sections even when bent about their strong axis. Figure 16.7, which is based on
elastic critical load theory analogous to the Euler buckling of struts, shows that
typical RHS beams will be of the order of ten times more stable than UB or UC
sections of the same area. The limits on l below which buckling will not affect Mb
of Table 38 of BS 5950: Part 1, are sufficiently high (l = 340, 225 and 170 for D/B
ratios of 2, 3 and 4, and py = 275 N/mm2) that only in very rare cases will lateral –
torsional buckling be a design consideration.
Situations in which the form of construction employed automatically provides
some degree of lateral restraint or for which a bracing system is to be used to
enhance a beam’s strength require careful consideration. The fundamental requirement of any form of restraint if it is to be capable of increasing the strength of the
main member is that it limits the buckling type deformations. An appreciation of
exactly how the main member would buckle if unbraced is a prerequisite for the
provision of an effective system. Since lateral – torsional buckling involves both
lateral deflection and twist, as shown in Fig. 16.4, either or both deformations may
be addressed. Clauses 4.3.2 and 4.3.3 of BS 5950: Part 1 set out the principles governing the action of bracing designed to provide either lateral restraint or torsional
restraint. In common with most approaches to bracing design these clauses assume
Steel Designers' Manual - 6th Edition (2003)
447
-.
z,,
/7
/1
II
P
0
0
0)
0
(-7,
-0
C
C-
--
0
C
Fig. 16.7
a-
0
N.)
0
0
0
P
0
N\
P
0
I
Mr to
ratio of
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
for
box section
II
I
/'l
b
Basic design
Effect of type of cross-section on theoretical elastic critical moment
that the restraints will effectively prevent movement at the braced cross-sections,
thereby acting as if they were rigid supports. In practice, bracing will possess a finite
stiffness. A more fundamental discussion of the topic, which explains the exact
nature of bracing stiffness and bracing strength, may be found in References 7 and
8. Noticeably absent from the code clauses is a quantitative definition of ‘adequate
stiffness’, although it has subsequently been suggested that a bracing system that is
25 times stiffer than the braced beam would meet this requirement. Examination
of Reference 7 shows that while such a check does cover the majority of cases, it is
still possible to provide arrangements in which even much stiffer bracing cannot
supply full restraint.
Steel Designers' Manual - 6th Edition (2003)
448
Beams
Table 16.7 Maximum values of lLT for
which pb = py for rolled
sections
py (N/mm2)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
245
265
275
325
340
365
415
430
450
Value of lLT
up to which pb = py
37
35
34
32
31
30
28
27
26
16.4 Lateral bracing
For design to BS 5950: Part 1, unless the engineer is prepared to supplement the code
rules with some degree of working from first principles, only restraints capable of
acting as rigid supports are acceptable. Despite the absence of a specific stiffness
requirement, adherence to the strength requirement together with an awareness that
adequate stiffness is also necessary, avoiding obviously very flexible yet strong
arrangements, should lead to satisfactory designs. Doubtful cases will merit examination in a more fundamental way.7,8 Where properly designed restraint systems are
used the limits on lLT for Mb = Mc (or more correctly pb = py) are given in Table 16.7.
For beams in plastically-designed structures it is vital that premature failure due
to plastic lateral – torsional buckling does not impair the formation of the full plastic
collapse mechanism and the attainment of the plastic collapse load. Clause 5.3.3
provides a basic limit on L/ry to ensure satisfactory behaviour; it is not necessarily
compatible with the elastic design rules of section 4 of the code since acceptable
behaviour can include the provision of adequate rotation capacity at moments
slightly below Mp.
The expression of clause 5.3.3 of BS 5950: Part 1,
38ry
Lm £
1
2 2
(16.4)
[ f /130 + (p / 275) (x / 36) ]
2
c
y
makes no allowance for either of two potentially beneficial effects:
(1) moment gradient
(2) restraint against lateral deflection provided by secondary structural members
attached to one flange as by the purlins on the top flange of a portal frame
rafter.
The first effect may be included in Equation (16.4) by adding the correction term
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Bracing action in bridges – U-frame design
449
of Brown,9 the basis of which is the original work on plastic instability of Horne.10
This is covered explicitly in clause 5.3.3. A method of allowing for both effects when
the beam segment being checked is either elastic or partially plastic is given in
Appendix G of BS 5950: Part 1; alternatively the effect of intermittent tension flange
restraint alone may be allowed for by replacing Lm with an enhanced value Ls
obtained from clause 5.3.4 of BS 5950: Part 1.
In both cases the presence of a change in cross-section, for example, as produced
by the type of haunch usually used in portal frame construction, may be allowed
for. When the restraint is such that lateral deflection of the beam’s compression
flange is prevented at intervals, then Equation (16.4) applies between the points
of effective lateral restraint. A discussion of the application of this and other
approaches for checking the stability of both rafters and columns in portal frames
designed according to the principles of either elastic or plastic theory is given in
section 18.7.
16.5 Bracing action in bridges – U-frame design
The main longitudinal beams in several forms of bridge construction will, by virtue
of the structural arrangement employed, receive a significant measure of restraint
against lateral – torsional buckling by a device commonly referred to as U-frame
action. Figure 16.8 illustrates the original concept based on the half-through girder
form of construction. (See Chapter 4 for a discussion of different bridge types.) In
a simply-supported span, the top (compression) flanges of the main girders, although
laterally unbraced in the sense that no bracing may be attached directly to them,
cannot buckle freely in the manner of Fig. 16.4 since their lower flanges are
restrained by the deck. Buckling must therefore involve some distortion of the
girder web into the mode given in Fig. 16.8 (assuming that the end frames prevent
lateral movement of the top flange). An approximate way of dealing with this is
to regard each longitudinal girder as a truss in which the tension chord is fully
compression boom
Fig. 16.8
Buckling of main beams of half-through girder
Steel Designers' Manual - 6th Edition (2003)
450
Beams
U—frame
(a)
unit
unit
load
load
(b)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 16.9 U-frame restraint action. (a) Components of U-frame. (b) U-frame elastic support
stiffness
laterally restrained and the web members, by virtue of their lateral bending stiffness, inhibit lateral movement of the top chord. It is then only a small step to regard
this top chord as a strut provided with a series of intermediate elastic spring
restraints against buckling in the horizontal plane. The stiffness of each support corresponds to the stiffness of the U-frame comprising the two vertical web stiffeners
and the cross-girder and deck shown in Fig. 16.9.
The elastic critical load for the top chord is
Pcr = p 2 EI / L2E
(16.5)
in which LE is the effective length of the strut.
If the strut receives continuous support of stiffness (1/d LR) per unit length, in
which LR is the distance between U-frame restraints, and buckles in a single halfwave, this load will be given by
Pcr = (p 2 EI / L2 ) + (L2 /p 2dLR )
(16.6)
which gives a minimum value when
L = p (EIdLR )
0.25
(16.7)
giving
Pcr = 2(EI /dLR )
0.5
(16.8)
or
LE = (p / ÷2)( EIdLR )
0.25
(16.9)
If lateral movement at the ends of the girder is not prevented by sufficiently
stiff end U-frames, the mode will be as shown in Fig. 16.10. The effective length is
then:
LE = p ( EIdLR )
0.25
In clause 9.6.4.1.1.2 of BS 5400: Part 3 the effective length is given by
LE = k( EI c LR d )
0.25
(16.10)
Steel Designers' Manual - 6th Edition (2003)
Bracing action in bridges – U-frame design
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 16.10
451
Buckling mode for half-through construction with flexible end frames
where K is a parameter that takes account of the stiffness of the end U-frames. For
effectively rigid frames, K = 2.22, which is the same as p/÷2.
For unstiffened girders a similar approach is possible with the effective U-frame
now comprising a unit length of girder web plus the cross-member. In all cases the
assessment of U-frame stiffness via the d parameter is based on summing the deflections due to bending of the horizontal and vertical components, including any flexibility of the upright to cross-frame connections. Clauses 9.6.4.1.3 and 9.6.4.2.2 deal
respectively with the cases where actual vertical members are either present or
absent.
Because the U-frames are required to resist the buckling deformations, they will
attract forces which may be estimated as the product of the additional deformation,
as a proportion of the initial lateral deformation of the top chord, and the U-frame
stiffness as
FR =
1
Ê
ˆ LE
-1
Ë 1 - s fc /s ci
¯ 667d
or FR =
Ê s fc ˆ LE
Ë s ci - s fc ¯ 667d
(16.11)
in which the assumed initial bow over an effective length of flange (LE) has been
taken as LE/667, and 1/(1 - sfc/sci) is the amplification, which depends in a non-linear
fashion on the level of stress sfc in the flange.
For a frame spacing LR and a flange critical stress corresponding to a force level
of p 2EIc/LR2, the maximum possible value of FR given in clause 9.12.2 of BS 5400:
Part 3 is
FR =
Ê s fc ˆ
EI / 16.7L2R
Ë s ci - s fc ¯ c
(16.12)
Additional forces in the web stiffeners are produced by rotation q of the ends of
the cross beam due to vertical loading on the cross beam. Clause 9.12.2.3 of BS 5400:
Part 3 evaluates the additional force as:
Fc = 3EI Iq / d22
(16.13)
Steel Designers' Manual - 6th Edition (2003)
452
Beams
Fc = qd
1
Ê
ˆ
Ë 1.5d + L3R / 12EI c ¯
(16.14)
In this expression, q is the difference in rotation between that at the U-frame and
the mean of the rotations at the adjacent frames on either side. The division in the
expression represents the combined flexibility of the frame (conservatively taken as
1.5d) and of the compression flange in lateral bending.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
16.6 Design for restricted depth
Frequently beam design will be constrained by a need to keep the beam depth to a
minimum. This restriction is easy to understand in the context of floor beams in a
multi-storey building for which savings in overall floor depth will be multiplied
several times over, thereby permitting the inclusion of extra floors within the same
overall building height or effecting savings on expensive cladding materials by
reducing building height for the same number of floors. Within the floor zone of
buildings with large volumes of cabling, ducting and other heavy services, only a
fraction of the depth is available for structural purposes.
Such restrictions lead to a number of possible solutions which appear to run contrary to the basic principles of beam design. However, structural designers should
remember that the main framing of a typical multi-storey commercial building
typically represents less than 10% of the building cost and that factors such as the
efficient incorporation of the services and enabling site work to proceed rapidly
and easily are likely to be of greater overall economic significance than trimming
steel weight.
An obvious solution is the use of universal columns as beams. While not as structurally efficient for carrying loads in simple vertical bending as UB sections, as illustrated by the example of Table 16.8, their design is straightforward. Problems of web
bearing and buckling at supports are less likely due to the reduced web d/t ratios.
Lateral–torsional buckling considerations are less likely to control the design of laterally unbraced lengths because the wider flanges will provide greater lateral
stiffness (L/ry values are likely to be low). Wider flanges are also advantageous for
supporting floor units, particularly the metal decking used frequently as part
of a composite floor system.
Difficulties can occur, because of the reduced depth, with deflections, although
dead load deflections may be taken out by precambering the beams. This will not
assist in limiting deflections in service due to imposed loading, although composite
action will provide a much stiffer composite section. Excessive deflection of the floor
beams under the weight of wet concrete can significantly increase slab depths at
mid-span, leading to a substantially higher dead load. None of these problems need
cause undue difficulty provided they are recognized and the proper checks made at
a sufficiently early stage in the design.
Another possible source of difficulty arises in making connections between
shallow beams and columns or between primary and secondary beams. The reduced
Steel Designers' Manual - 6th Edition (2003)
Design for restricted depth
453
Table 16.8 Comparison of use of UB and UC for simple beam
design
71 kN/m
6m
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Mmax = 320 kN m
Fv = 213 kN
beams at 3 m spacing
457 ¥ 152 ¥ UB 60
254 ¥ 254 ¥ UC 89
Mc = 352 kN m
Pv = 600 kN
Fv < 0.6 Pv - no interaction
Mc = 326 kN m
Pv = 435 kN
Fv < 0.6 Pv - no interaction
From Equation (16.2) and Table 16.5, assuming deflection limit
is L/360 and service load is 47 kN m
Irqd = 2.23 (47 ¥ 3) 62 = 11 319 cm4
Ix = 25 500 cm4
Ix = 14 300 cm4
web depth can lead to problems in physically accommodating sufficient bolts to
carry the necessary end shears. Welding cleats to beams removes some of the dimensional tolerances that assist with erection on site as well as interfering with the
smooth flow of work in a fabricator’s shop that is equipped with a dedicated saw
and drill line for beams. Extending the connection beyond the beam depth by using
seating cleats is one solution, although a requirement to contain the connection
within the beam depth may prevent their use.
Beam depths may also be reduced by using moment-resisting beam-to-column
connections which provide end fixity to the beams; a fixed end beam carrying a
central point load will develop 50% of the peak moment and only 20% of the central
deflection of a similar simply-supported beam. Full end fixity is unlikely to be a realistic proposition in normal frames but the replacement of the notionally pinned
beam-to-column connection provided by an arrangement such as web cleats, with a
substantial end plate that functions more or less as a rigid connection, permits the
development of some degree of continuity between beams and columns. These
arrangements will need more careful treatment when analysing the pattern of internal moments and forces in the frame since the principles of simple construction will
no longer apply.
An effect similar to the use of UC sections may be achieved if the flanges of a
UB of a size that is incapable of carrying the required moment are reinforced by
welding plates over part of its length. Additional moment capacity can be provided
where it is needed as illustrated in Fig. 16.11; the resulting non-uniform section is
stiffer and deflects less. Plating of the flanges will not improve the beam’s shear
capacity since this is essentially provided by the web and the possibility of shear
or indeed local web capacity governing the design must be considered. A further
Steel Designers' Manual - 6th Edition (2003)
454
Beams
4,'
for base section
M for compound section
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 16.11
Selective increase of moment capacity by use of a plated UB
development of this idea is the use of tapered sections fabricated from plate.11 To
be economic, tapered sections are likely to contain plate elements that lie outside
the limits for compact sections.
Because of the interest in developing longer spans for floors and the need to
improve the performance of floor beams, a number of ingenious arrangements have
developed in recent years.12 Since these all utilize the benefits of composite action
with the floor slab, they are considered in Chapter 21.
16.7 Cold-formed sections as beams
In situations where a relatively lightly loaded beam is required such as a purlin or
sheeting rail spanning between main frames supporting the cladding in a portal
frame, it is common practice to use a cold-formed section produced cold from flat
steel sheet, typically between about 1 mm and 6 mm in thickness, in a wide range of
shapes of the type shown in Fig. 16.12. A particular feature is that normally each
section is formed from a single flat bent into the required shape; thus most available sections are not doubly symmetric but channels, zeds and other singly symmetric shapes. The forming process does, however, readily permit the use of quite
complex cross-sections, incorporating longitudinal stiffening ribs and lips at the
edges of flanges. Since the original coils are usually galvanized, the members do not
normally require further protective treatment.
The structural design of cold-formed sections is covered by BS 5950: Part 5, which
permits three approaches:
(1) design by calculation using the procedures of the code, section 5, for members
in bending
(2) design on the basis of testing using the procedures of section 10 to control the
testing and section 10.3 for members in bending
(3) for three commonly used types of member (zed purlins, sheeting rails and lattice
joists), design using the simplified set of rules given in section 9.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Cold-formed section as beams
Fig. 16.12
455
Typical cold-formed section beam shapes
In practice option (2) is the most frequently used, with all the major suppliers providing design literature, the basis of which is usually extensive testing of their
product range, design being often reduced to the selection of a suitable section for
a given span, loading and support arrangement using the tables provided.
Most cold-formed section types are the result of considerable development work
by their producers. The profiles are therefore highly engineered so as to produce a
Steel Designers' Manual - 6th Edition (2003)
456
Beams
—-i
distribution of
compressive stress
at progressively
J higher levels of
applied load
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 16.13
Loss of plating effectiveness at progressively higher compressive stress
near optimum performance, a typical example being the ranges of purlins produced
by the leading UK suppliers. Because of the combination of the thin material and
the comparative freedom provided by the forming process, this means that most sections will contain plate elements having high width-to-thickness ratios. Local buckling effects, due either to overall bending because the profile is non-compact, or to
the introduction of localized loads, are of greater importance than is usually the case
for design using hot-rolled sections. BS 5950: Part 5 therefore gives rather more
attention to the treatment of slender cross-sections than does BS 5950: Part 1. In
addition, manufacturers’ design data normally exploit the post-buckling strength
observed in their development tests.
The approach used to deal with sections containing slender elements in BS 5950:
Part 5 is the well accepted effective width technique. This is based on the observation that plates, unlike struts, are able to withstand loads significantly in excess of
their initial elastic buckling load, provided some measure of support is available to
at least one of their longitudinal edges. Buckling then leads to a redistribution of
stress, with the regions adjacent to the supported longitudinal edges attracting
higher stresses and the other parts of the plate becoming progressively less effective, as shown in Fig. 16.13. A simple design representation of the condition of
Fig. 16.13 consists of replacing the actual post-buckling stress distribution with the
approximation shown in Fig. 16.14. The structural properties of the member
Fig. 16.14
Effective width design approximation
Steel Designers' Manual - 6th Edition (2003)
Beams with web openings
be/2
__
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 16.15
457
be/2
I -1
Effective cross-section
(strength and stiffness) are then calculated for this effective cross-section as illustrated in Fig. 16.15. Tabulated information in BS 5950: Part 5 for steel of yield
strength 280 N/mm2 makes the application of this approach simpler in the sense that
effective widths may readily be determined, although cross-sectional properties
have still to be calculated. The use of manufacturer’s literature removes this requirement. For beams, Part 5 also covers the design of reinforcing lips on the usual basis
of ensuring that the free edge of a flange supported by a single web behaves as if
both edges were supported; web crushing under local loads, lateral–torsional buckling and the approximate determination of deflections take into account any loss of
plating effectiveness.
For zed purlins or sheeting rails section 9 of BS 5950: Part 5 provides a set of
simple empirically based design rules. Although easy to use, these are likely to lead
to heavier members for a given loading, span and support arrangement than either
of the other permitted procedures. A particular difference of this material is its use
of unfactored loads, with the design conditions being expressed directly in member
property requirements.
16.8 Beams with web openings
One solution to the problem of accommodating services within a restricted floor
depth is to run the services through openings in the floor beams. Since the size of
hole necessary in the beam web will then typically represent a significant proportion of the clear web depth, it may be expected that it will have an effect on structural performance. The easiest way of visualizing this is to draw an analogy between
a beam with large rectangular web cut-outs and a Vierendeel girder. Figure 16.16
shows how the presence of the web hole enables the beam to deform locally in a
similar manner to the shear type deformation of a Vierendeel panel. These deformations, superimposed on the overall bending effects, lead to increased deflection
and additional web stresses.
A particular type of web hole is the castellation formed when a UB is cut, turned
and rewelded as illustrated in Fig. 16.17. For the normal UK module geometry this
leads to a 50% increase in section depth with a regular series of hexagonal holes.
Other geometries are possible, including a further increase in depth through the use
Steel Designers' Manual - 6th Edition (2003)
458
Beams
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(a)
(b)
Fig. 16.16 Vierendeel-type action in beam with web openings: (a) overall view, (b) detail of
deformed region
bottom tee
(0)
b 2o b 2a
(b)
Fig. 16.17
Castellated beam: (a) basic concept, (b) details of normal UK module geometry
Steel Designers' Manual - 6th Edition (2003)
Beams with web openings
459
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
of plates welded between the two halves of the original beam. Some aspects of the
design of castellated beams are covered by the provisions of BS 5950: Part 1, while
more detailed guidance is available in a Constrado publication.13
Based on research conducted in the USA, a comparatively simple elastic method
for the design of beams with web holes, including a fully worked example, is available.14 This uses the concept of an analysis for girder stresses and deflections that
neglects the effects of the holes, coupled with checking against suitably modified
limiting values. The full list of design checks considered in Reference 14 is:
(1) web shear due to overall bending acting on the reduced web area
(2) web shear due to local Vierendeel bending at the hole
(3) primary bending stresses (little effect since overall bending is resisted principally by the girder flanges)
(4) local bending due to Vierendeel action
(5) local buckling of the tee formed by the compression flange and the web adjoining the web hole
(6) local buckling of the stem of the compression tee due to secondary bending
(7) web crippling under concentrated loads or reactions near a web hole; as a simple
guide, Reference 14 suggests that for loads which act at least (d/2) from the edge
of a hole this effect may be neglected
(8) shear buckling of the web between holes; as a simple guide, Reference 14 suggests that for a clear distance between holes that exceeds the hole length this
effect may be neglected
(9) vertical deflections; as a rough guide, secondary effects in castellated beams may
be expected to add about 30% to the deflections calculated for a plain web beam
of the same depth (1.5D). Beams with circular holes of diameter (D/2) may be
expected to behave similarly, while beams with comparable rectangular holes
may be expected to deflect rather more.
As an alternative to the use of elastic methods, significant progress has been made
in recent years in devising limit state approaches based on ultimate strength
conditions. A CIRIA/SCI design guide15 dealing with the topic principally from the
point of composite beams is now available. If some of the steps in the 24-point design
check of Reference 15 are omitted, the method may be applied to non-composite
beams, including composite beams under construction. Much of the basis for Reference 15 may be traced back to the work of Redwood and Choo,16 and the following treatment of bare steel beams is taken from Reference 16.
The governing condition for a stocky web in the vicinity of a hole is taken as
excessive plastic deformation near the opening corners and in the web above and
below the opening as illustrated in Fig. 16.18. A conservative estimate of web
strength may then be obtained from a moment–shear interaction diagram of the
type shown as Fig. 16.19. Values of M0 and V1 in terms of the plastic moment capacity and plastic shear capacity of the unperforated web are given in Reference 14 for
both plain and reinforced holes; M1 may also be determined in this way. Solution of
these equations is tedious, but some rearrangement and simplification are possible
Steel Designers' Manual - 6th Edition (2003)
460
Beams
cracking
concrete crushing
yielding
support
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 16.18
Hole-induced failure
(V\12
v11
+
/ M—N1
2
) = 1.0
M0—U1
V1
shear
Fig. 16.19
Moment–shear interaction for a stocky web in the vicinity of a hole
so that an explicit solution for the required area of reinforcement may be obtained.
However, the whole approach is best programmed for a microcomputer, and a
program based on the full method of Reference 15 is available from the SCI.
References to Chapter 16
1. British Standards Institution (2000) Part 1: Code of practice for design in simple
and continuous construction: hot rolled sections. BS 5950, BSI, London.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
References
461
2. The Steel Construction Institute (SCI) (2001) Steelwork Design Guide to BS
5950: Part 1: 2000, Vol. 1, Section Properties, Member Capacities, 6th edn. SCI,
Ascot, Berks.
3. Johnson R.P. & Buckby R.J. (1979) Composite Structures of Steel and Concrete,
Vol. 2: Bridges with a Commentary on BS 5400: Part 5, 1st edn. Granada, London.
(2nd edn, 1986).
4. Woodcock S.T. & Kitipornchai S. (1987) Survey of deflection limits for portal
frames in Australia. J. Construct. Steel Research, 7, No. 6, 399–418.
5. Nethercot D.A., Salter P. & Malik A. (1989) Design of Members Subject to
Bending and Torsion. The Steel Construction Institute, Ascot, Berks (SCI Publication 057).
6. Dux P.F. & Kitipornchai S. (1986) Elastic buckling strength of braced beams.
Steel Construction, (AISC), 20, No. 1, May.
7. Trahair N.S. & Nethercot D.A. (1984) Bracing requirements in thin-walled structures. In Developments in Thin-Walled Structures – 2 (Ed. by J. Rhodes & A.C.
Walker), pp. 93–130. Elsevier Applied Science Publishers, Barking, Essex.
8. Nethercot D.A. & Lawson R.M. (1992) Lateral stability of steel beams and
columns – common cases of restraint. SCI Publication 093. The Steel Construction Institute, Ascot, Berks.
9. Brown B.A. (1988) The requirements for restraints in plastic design to BS 5950.
Steel Construction Today, 2, No. 6, Dec., 184–6.
10. Horne M.R. (1964) Safe loads on I-section columns in structures designed by
plastic theory. Proc. Instn. Civ. Engrs, 29, Sept., 137–50.
11. Raven G.K. (1987) The benefits of tapered beams in the design development of
modern commercial buildings. Steel Construction Today, 1, No. 1, Feb., 17–25.
12. Owens G.W. (1987) Structural forms for long span commercial building and
associated research needs. In Steel Structures, Advances, Design and Construction (Ed. by R. Narayanan), pp. 306–319. Elsevier Applied Science Publishers,
Barking, Essex.
13. Knowles P.R. (1985) Design of Castellated Beams for use with BS 5950 and BS
449. Constrado.
14. Constrado (1977) Holes in Beam Webs: Allowable Stress Design. Constrado.
15. Lawson R.M. (1987) Design for Openings in the Webs of Composite Beams.
CIRIA Special Publication S1 and SCI Publication 068. CIRIA/Steel Construction Institute.
16. Redwood R.G. & Choo S.H. (1987) Design tools for steel beams with web openings. In: Composite Steel Structures, Advances, Design and Construction (Ed. by
R. Narayanan), pp. 75–83. Elsevier Applied Science Publishers, Barking, Essex.
A series of worked examples follows which are relevant to Chapter 16.
Steel Designers' Manual - 6th Edition (2003)
462
Worked examples
The
Steel Construction
Institute
Subject
Chapter ref.
BEAM EXAMPLE 1
LATERALLY RESTRAINED
UNIVERSAL BEAM
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
Made by
DAN
16
Sheet no.
1
Checked by GWO
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Problem
Select a suitable UB section to function as a simply supported beam
carrying a 140 mm thick solid concrete slab together with an
imposed load of 7.0 kN/m2. Beam span is 7.2 m and beams are
spaced at 3.6 m intervals. The slab may be assumed capable of providing continuous lateral restraint to the beam’s top flange.
7.2 in
Due to restraint from slab there is no possibility of lateral-torsional
buckling, so design beam for:
i)
ii)
iii)
Moment capacity
Shear capacity
Deflection limit
Loading
D .L . =
( 2.4 ¥ 9.81 ¥ 0.14 )
= 3.3 kN / m2
I.L.
= 7.0 kN/m2
Total serviceability loading
= 10.3 kN/m2
Total load for ultimate limit state
= 1.4 ¥ 3.3 + 1.6 ¥ 7.0
= 15.8 kN/m2
Design ultimate moment
= (15.8 ¥ 3.6 ) ¥ 7.22/8
= 369 kN m
Design ultimate shear
= (15.8 ¥ 3.6) ¥ 7.2/2
= 205 kN
Table 2
Steel Designers' Manual - 6th Edition (2003)
Worked examples
Subject
The
Steel Construction
Institute
Chapter ref.
BEAM EXAMPLE 1
LATERALLY RESTRAINED
UNIVERSAL BEAM
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
Made by
Sheet no.
2
Checked by GWO
Assuming use of S275 steel and no material greater than
16 mm thick,
Table 9
= 275 N/mm2
take py
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
DAN
16
Required Sx = 369 ¥ 106/275
= 1.34 ¥ 106 mm3 = 1340 cm3
A 457 ¥ 152 ¥ 67 UB has a value of Sx of 1440 cm3
T = 15.0 < 16.0 mm
Steelwork
Design Guide
Vol 1
\ py = 275 N/mm2
Check section classification
Actual b/T = 5.06
1/2
e = (275/py)
3.5.2
d/t = 44.7
=1
Table 11
Limit on b/T for plastic section = 9 > 5.06
Limit on d/t for shear = 63 > 44.7
\ Section is plastic
Actual
Mc = 275 ¥ 1440 ¥ 103
4.2.5
= 396 ¥ 106 N mm
= 396kN m > 369kN m OK
Vertical shear capacity
Pv = 0.6 pyAv
where
Av = tD
\
Pv = 0.6 ¥ 275 ¥ 9.1 ¥ 457.2 = 686 ¥ 103 N
= 686kN > 205kN OK
4.2.3
463
Steel Designers' Manual - 6th Edition (2003)
Worked examples
464
Subject
The
Steel Construction
Institute
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
Check serviceability deflections under imposed load
5 ¥ (7.0 ¥ 3.6 ) ¥ 7200
384 ¥ 205000 ¥ 32400 ¥ 10 4
= 13.3mm = span / 541
4
d=
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Chapter ref.
BEAM EXAMPLE 1
LATERALLY RESTRAINED
UNIVERSAL BEAM
From Table 5, limit is span/360
\ d OK
\ Use 457 ¥ 152 ¥ 67UB Grade 43
Made by
DAN
16
Sheet no.
3
Checked by GWO
2.5.1
Steel Designers' Manual - 6th Edition (2003)
Worked examples
The
Steel Construction
Institute
Subject
Chapter ref.
BEAM EXAMPLE 2
LATERALLY UNRESTRAINED
UNIVERSAL BEAM
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
Made by
DAN
16
Sheet no.
1
Checked by GWO
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Problem
For the same loading and support conditions of example 1 select a
suitable UB assuming that the member must be designed as laterally
unrestrained.
It is not now possible to arrange the calculations in such a way that
a direct choice is possible; a guess and check approach must be
adopted.
Try
610 ¥ 229 ¥ 125 UB
u
= 0.873
ry = 4.98 cm
x
= 34.0
Sx = 3680 cm3
l
= LE / ry = 7200 / 49.8
= 145
Steelwork
Design Guide
Vol 1
4.3.7.5
l/x = 145/34 = 4.26
v
= 0.85
Table 19
l LT = u v l
\ l LT = 0.873 ¥ 0.85 ¥ 145
= 108
Pb
= 116 N/mm2
Mb
= Sxpb = 3680 ¥ 103 ¥ 116
= 427 ¥ 106 N mm
= 427kN m > 369kN m OK
Table 20
4.3.7.3
465
Steel Designers' Manual - 6th Edition (2003)
466
Worked examples
Subject
The
Steel Construction
Institute
Chapter ref.
BEAM EXAMPLE 2
LATERALLY UNRESTRAINED
UNIVERSAL BEAM
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
Made by
DAN
16
Sheet no.
2
Checked by GWO
Since section is larger than before, Pv and d will also be satisfactory
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
\ Adopt 610 ¥ 229 ¥ 125UB
However, for a UDL, MLT = 0.925
Refer to member capacities section for values of Mb directly, noting
S275 material.
Table 18
Steelwork
Design Guide
Vol 1
Value of Mb required is 369 kNm; interpolating for MLT = 0.925 &
LE = 7.2 m suggests as possible sections:
533 ¥ 210 ¥ 122 UB
Mb OK
610 ¥ 229 ¥ 113 UB
Mb OK
Since both are larger than that checked for shear capacity and serviceability
deflection in the previous example, either may be adopted.
\
Adopt 533 ¥ 210 ¥ 122UB or 610 ¥ 229 ¥ 113UB Grade S275
Steel Designers' Manual - 6th Edition (2003)
Worked examples
Subject
The
Steel Construction
Institute
Chapter ref.
BEAM EXAMPLE 3
UNIVERSAL BEAM
SUPPORTING POINT LOADS
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
Made by
DAN
Checked by GWO
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Problem
Select a suitable UB section in S275 steel to carry the pair of point
loads at the third points transferred by crossbeams as shown in the
accompanying sketch. Design to BS 5950: Part 1.
kiV
4,116
A
3O,n
D
3.Om
3.Onz
>(
The crossbeams may reasonably be assumed to provide full lateral
and torsional restraint at B and C; assume further that ends A and
D are similarly restrained. Thus the actual level of transfer of load
at B and C (relative to the main beam’s centroid) will have no effect,
the lateral-torsional buckling aspects of the design being one of considering the 3 segments AB, BC and CD separately.
From statics the BMD and SFD are
Vm
BMD
406 kNm
SFD
126 kN
135 kN
16
Sheet no.
1
467
Steel Designers' Manual - 6th Edition (2003)
468
Worked examples
Subject
The
Steel Construction
Institute
Silwood Park, Ascot, Berks SL5 7QN
Chapter ref.
BEAM EXAMPLE 3
UNIVERSAL BEAM
SUPPORTING POINT LOADS
Design code
BS 5950: Part 1
Made by
DAN
16
Sheet no.
2
Checked by GWO
For initial trial section select a UB with Mcx > 406 kNm.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
A 457 ¥ 152 ¥ 74 UB provides Mcx of 429 kNm. Now check
lateral-torsional buckling strength for segments AB, BC & CD.
Steelwork
Design Guide
Vol 1
AB
b
= 0/406
= 0.0
mLT
= 0.57
—
M
= 0.57 ¥ 406 = 231.4 kNm
Table 18
For LE = 3.0 m, Mb = 288 kNm
\ 288 > 231.4 kNm OK
Steelwork
Design Guide
Vol 1
BC
b
= 377/406
mLT
—
M
= 0.97
= 0.93
Table 18
= 0.97 ¥ 406 = 393.8 kNm
But for LE = 3.0 m,
Mb = 288 kNm Not OK
Try 457 ¥ 191 ¥ 82 UB
this provides Mb of 396 kNm > 393.8 kNm OK
CD
Satisfactory by inspection OK
A 457 ¥ 191 ¥ 82 UB provides sufficient resistance to lateraltorsional buckling for each segment and thus for the beam as a
whole.
Steelwork
Design Guide
Vol 1
Steel Designers' Manual - 6th Edition (2003)
Worked examples
The
Steel Construction
Institute
Silwood Park, Ascot, Berks SL5 7QN
Subject
Chapter ref.
BEAM EXAMPLE 3
UNIVERSAL BEAM
SUPPORTING POINT LOADS
Design code
BS 5950: Part 1
Made by
DAN
16
Sheet no.
3
Checked by GWO
Check shear capacity; maximum shear at L is 135 kN
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Pv = 752 kN > 135 kN OK
Check bearing and buckling capacity of web at the supports –
required capacity is 135 kN.
Since C1 values exceed 135 kN in both cases, section is clearly
adequate.
Steelwork
Design Guide
Vol 1
For initial check on serviceability deflections assume as equivalent
UDL and factor down all loads by 1.5 to obtain
w=
145 + 116
= 19.3 kN /m
1.5 ¥ 9
5 ¥ 19.3 ¥ 9000 4
384 ¥ 205000 ¥ 37100 ¥ 10 4
= 21.68 mm = span/ 415
d=
Limiting deflection is span/360 \ OK
Beam is clearly satisfactory for deflection since these (approximate)
calculations have used the full load and not just the imposed load.
\ Adopt 457 ¥ 191 ¥ 82UB
Table 5
469
Steel Designers' Manual - 6th Edition (2003)
Chapter 17
Plate girders
by TERENCE M. ROBERTS and RANGACHARI NARAYANAN
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
17.1 Introduction
Plate girders are employed to support heavy vertical loads over long spans for which
the resulting bending moments are larger than the moment resistance of available
rolled sections. In its simplest form the plate girder is a built-up beam consisting of
two flange plates, fillet welded to a web plate to form an I-section (see Fig. 17.1).
The primary function of the top and bottom flange plates is to resist the axial compressive and tensile forces caused by the applied bending moments; the main function of the web is to resist the shear. Indeed this partition of structural action is used
as the basis for design in some codes of practice.
For a given bending moment the required flange areas can be reduced by increasing the distance between them. Thus for an economical design it is advantageous to
increase the distance between flanges. To keep the self-weight of the girder to a
minimum the web thickness should be reduced as the depth increases, but this leads
to web buckling considerations being more significant in plate girders than in rolled
beams.
Plate girders are sometimes used in buildings and are often used in small to
medium span bridges. They are designed in accordance with the provisions contained in BS 5950: Part 1: 20001 and BS 5400: Part 32 respectively. This chapter
explains current practice in designing plate girders for buildings and bridges; references to the relevant clauses in the codes are made.
17.2 Advantages and disadvantages
The development of highly automated workshops in recent years has reduced the
fabrication costs of plate girders very considerably; box girders and trusses still have
to be fabricated manually, with consequently high fabrication costs. Optimum use
of material is made, compared with rolled sections, as the girder is fabricated from
plates and the designer has greater freedom to vary the section to correspond with
changes in the applied forces. Thus variable depth plate girders have been increasingly designed in recent years. Plate girders are aesthetically more pleasing than
trusses and are easier to transport and erect than box girders.
470
Steel Designers' Manual - 6th Edition (2003)
Initial choice of cross-section for plate girders in buildings
471
transverse
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
stiffener
fitlet
weld
—fl.--d
A—A
N
Fig. 17.1
Elevation and cross-section of a typical plate girder
There are only a very few limitations in the use of plate girders. Compared with
trusses they are heavier, more difficult to transport and have larger wind resistance.
The provision of openings for services is also more difficult. The low torsional stiffness of plate girders makes them difficult to use in bridges having small plan radius.
Plate girders can sometimes pose problems during erection because of concern for
the stability of compression flanges.
17.3 Initial choice of cross-section for plate girders in buildings
17.3.1 Span-to-depth ratios
Advances in fabrication methods allow the economic manufacture of plate girders
of constant or variable depth. Traditionally, constant-depth girders were more
common in buildings; however, this may change as designers become more inclined
to modify the steel structure to accommodate services.3 Recommended span-todepth ratios are given in Table 17.1.
17.3.2 Recommended plate thickness and proportions
In general the slenderness of the cross-sections of plate girders used in buildings
should not exceed the limits specified for class 3 semi-compact cross-sections (clause
Steel Designers' Manual - 6th Edition (2003)
472
Plate girders
Table 17.1 Recommended span-to-depth ratios for plate girders used in buildings
Applications
Span-to-depth ratio
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(1) Constant-depth beams used in simply-supported composite girders, and
for simply-supported non-composite girders, with concrete decking
(2) Constant-depth beams used in continuous non-composite girders using
concrete decking (NB continuous composite girders are rare in buildings)
(3) Simply-supported crane girders (non-composite construction is usual)
12 to 20
15 to 20
10 to 15
3.5 of the Code), even though more slender cross-sections are permitted. The choice
of plate thickness is related to buckling. If the plates are too thin they may require
stiffening to restore adequate stiffness and strength, and the extra workmanship
required is expensive.
In view of the above the maximum depth-to-thickness ratio (d/t) of the webs of
plate girders in buildings is usually limited to
Ê 275 ˆ
d t < 120 e = 120Á
˜
Ë pyw ¯
1 2
where pyw is the design strength of the web plate. The outstand width-to-thickness
ratio of the compression flange (b/T) is usually limited to
Ê 275 ˆ
b T £ 13 e = 13Á
˜
Ë pyf ¯
1 2
where pyf is the design strength of the compression flange.
Changes in flange size along the girder are not usually worthwhile in buildings.
For non-composite girders the flange width is usually within the range 0.3–0.5 times
the depth of the section (0.4 is most common). For simply-supported composite
girders these guidelines can still be employed for preliminary sizing of compression
flanges. The width of tension flanges can be increased by 30%.
17.3.3 Stiffeners
Horizontal web stiffeners are not usually required for plate girders used in buildings. Vertical web stiffeners may be provided to enhance the resistance to shear near
the supports. Intermediate stiffening at locations far away from supports will, in
general, be unnecessary due to reduced shear.
The provision of vertical or transverse web stiffeners increases both the critical
shear strength qcr (initial buckling strength) and the shear buckling strength qw
(post-buckling strength) of web panels. The critical shear strength is increased by
a reduction in the web panel aspect ratio a/d (width/depth). Shear buckling strength
is increased by enhanced tension field action, whereby diagonal tensile membrane
Steel Designers' Manual - 6th Edition (2003)
Design of plate girders used in buildings to BS 5950: Part 1: 2000
473
stresses, which develop during the post-buckling phase, are resisted by the boundary members (vertical stiffeners and flanges).
Transverse stiffeners are usually spaced such that the web panel aspect ratio is
between 1.0 and 2.0, since there is little increase in strength for larger panel aspect
ratios. For end panels designed without utilizing tension field action the aspect ratio
is reduced to 0.6–1.0. Sometimes double stiffeners are employed as bearing stiffeners at the end supports, to form what is known as an end post. The overhang of the
girder beyond the support is generally limited to a maximum of one eighth of the
depth of the girder.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
17.4 Design of plate girders used in buildings to
BS 5950: Part 1: 2000
17.4.1 General
Any cross-section of a plate girder will normally be subjected to a combination of
shear force and bending moment, present in varying proportions. BS 5950: Part 1:
20001 specifies that the design of plate girders should satisfy the relevant provisions
given in clauses 4.2 (members subject to bending) and 4.3 (lateral-torsional buckling) together with the additional provisions in 4.4 (plate girders). The additional
provisions in clause 4.4 of the Code are related primarily to the susceptibility of
slender web panels to local buckling.
17.4.2 Dimensions of webs and flanges
Minimum web thickness requirements (clause 4.4.3 of the Code) are based on serviceability considerations, such as adequate stiffness to prevent unsightly buckles
developing during erection and in service, and also to avoid the compression flange
buckling into the web.
The buckling resistance of slender webs can be increased by the provision of web
stiffeners. In general the webs of plate girders used in buildings are either unstiffened or have transverse stiffeners only (see Fig. 17.1).
The following minimum web thickness values are prescribed to avoid serviceability problems.
d
250
(2) Transversely stiffened webs,
(1) Unstiffened webs,
t≥
For a > d
t≥
d
250
For a £ d
t≥
Ê d ˆÊ a ˆ
Ë 250 ¯ Ë d ¯
1 2
Steel Designers' Manual - 6th Edition (2003)
474
Plate girders
The following minimum web thickness values are prescribed to avoid the compression flange buckling into the web.
(3) Unstiffened webs
t≥
Ê d ˆ Ê pyf ˆ
Á
˜
Ë 250 ¯ Ë 345 ¯
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(4) Transversely stiffened webs
For a > 1.5d
t≥
Ê d ˆ Ê pyf ˆ
Á
˜
Ë 250 ¯ Ë 345 ¯
For a £ 1.5d
t≥
Ê d ˆ Ê pyf ˆ
Á
˜
Ë 250 ¯ Ë 455 ¯
1 2
Local buckling of the compression flange may also occur if the flange plate is of
slender proportions. In general there is seldom good reason for the b/T ratio of the
compression flanges of plate girders used in buildings to exceed the class 3 semicompact limit (clause 3.5 of the Code) b/T £ 13 e.
17.4.3 Moment resistance
17.4.3.1 Web not susceptible to shear buckling
Determination of the moment resistance Mc of laterally restrained plate girders
depends upon whether or not the web is susceptible to shear buckling. If the web
depth to thickness ratio d/t £ 62 e the web should be assumed not to be susceptible
to shear buckling, and the moment resistance of the section should be determined
in accordance with clause 4.2.5 of the Code.
17.4.3.2 Web susceptible to shear buckling
If the web depth to thickness ratio d/t > 62 e it should be assumed susceptible to
shear buckling. The moment resistance of the section Mc should be determined
taking account of the interaction of shear and moment, using the following methods.
(1) Low shear
Provided that the applied shear force Fv £ 0.6 Vw, where Vw is the simple shear buckling resistance from clause 4.4.5.2 of the Code (see section 17.4.4.2 (1)), the moment
resistance should be determined from clause 4.2.5 of the Code. For class 1 plastic
and class 2 compact cross-sections:
Mc = pyS
Steel Designers' Manual - 6th Edition (2003)
Design of plate girders used in buildings to BS 5950: Part 1: 2000
475
where py is the design strength of the steel and S is the plastic modulus. For class 3
semi-compact cross-sections:
Mc = pyZ
where Z is the elastic section modulus. For class 4 slender cross-sections:
Mc = pyZeff
where Zeff is the effective section modulus determined in accordance with clause
3.6.2 of the Code.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(2) High shear flanges-only method
If Fv > 0.6 Vw but the web is designed for shear only, provided that the flanges are
not class 4, a conservative value Mf for the moment resistance may be obtained by
assuming that the moment is resisted by the flanges only. Hence:
Mf = pyf Sf
where Sf is the plastic modulus of the flanges only.
(3) High shear general method
If Fv > 0.6 Vw and the applied moment does not exceed the low shear value given
by (1), the web should be designed using Annex H.3 of the Code for the applied
shear combined with any additional moment beyond the flanges-only moment
resistance Mf given by (2).
17.4.4 Shear resistance
17.4.4.1 Web not susceptible to shear buckling
If the web depth to thickness ratio d/t £ 62 e it is not susceptible to shear buckling
and the shear resistance Pv should be determined in accordance with clause 4.2.3 of
the Code, i.e.
Pv = 0.6pyw Av = 0.6pywtd
where Av = td is the shear area.
17.4.4.2 Web susceptible to shear buckling
If d/t > 62 e the shear buckling resistance of the web should be determined in accordance with clause 4.4.5 of the Code.
Steel Designers' Manual - 6th Edition (2003)
Plate girders
476
\
\\
/
q
06p
yw
:
P
old 20
I
I
I
I
I
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
100
Fig. 17.2
d/ t
200
Relationship between shear buckling strength qw and critical shear strength qcr
The procedures for determining the shear buckling resistance of slender plate
girder webs, given in BS 5950: Part 1: 2000,1 have been updated and are now consistent with the procedures given in Eurocode 3: Design of steel structures.4 Two
methods are specified, namely the simplified method and the more exact method.
Both methods are based on the post-critical shear buckling strength of the web qw,
and result in significantly greater values of shear resistance than the elastic critical
method of BS 5950: Part 1: 1985. The relationship between qw and qcr , the elastic
critical shear strength, is illustrated in Fig. 17.2. The more exact method incorporates
a flange related component of shear resistance and is comparable with the method
based on tension field theory in BS 5950: Part 1: 1985. The critical shear buckling
resistance Vcr, based on qcr, is retained only as a reference value and for the design
of particular types of end panel.
The two methods for determining the shear buckling resistance are as follows.
(1) Simplified method
The shear buckling resistance Vb of a web, with or without intermediate transverse
stiffeners, may be taken as the simple shear buckling resistance Vw given by:
Vw = dtqw
where qw is the shear buckling strength of the web. Values of qw are tabulated in
Table 21 of the Code for web panel aspect ratios from 0.4 to infinity. The equations
on which qw is based are specified in Annex H.1 of the Code.
(2) More exact method
Alternatively the shear buckling resistance Vb of a web panel between two transverse stiffeners may be determined as follows. If the flanges of the panel are fully
Steel Designers' Manual - 6th Edition (2003)
Design of plate girders used in buildings to BS 5950: Part 1: 2000
477
stressed, i.e. the mean longitudinal stress in the smaller flange due to moment or
axial force ff is equal to the flange strength pyf, then:
Vb = Vw = dtqw
If the flanges are not fully stressed ( ff < pyf):
Vb = Vw + Vf but
Vb £ Pv = 0.6pywtd
in which Vf is the flange-dependent shear buckling resistance given by
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Vf =
[
Pv (d a) 1 - ( ff pyf )
2
]
1 + 0.15(M pw M pf )
Mpf is the plastic moment resistance of the smaller flange about its own equal area
axis, perpendicular to the plane of the web. Mpw is the plastic moment resistance of
the web about its equal area axis perpendicular to the plane of the web, determined
using pyw. For a rectangular flange and a web of uniform thickness:
M pf = 0.25 pyf (2b + t )T 2
M pw = 0.25 pyw td 2
(3) Critical shear buckling resistance
The critical shear buckling resistance Vcr of the web of an I-section is given by
Vcr = dtqcr
where qcr is the critical shear buckling strength determined in accordance with
clause 4.4.5.4 of the Code as follows (see also Annex H.2 of the Code).
If Vw = Pv
If Pv > Vw > 0.72 Pv
then Vcr = Pv
then Vcr = (9 Vw - 2 Pv ) 7
If Vw £ 0.72 Pv
Ê Vw ˆ
then Vcr =
Ë 0.9 ¯
2
Pv
17.4.4.3 Panels with openings
Web openings frequently have to be provided in girders used in building construction for service ducts etc. When any dimension of such an opening exceeds 10% of
the minimum dimensions of the panel in which it is located, reference should be
made to clause 4.15 of the Code. Panels with openings should not be used as anchor
panels, and the adjacent panels should be designed as end panels.
Guidance on the design of plate girders with openings is provided in Reference 5.
Steel Designers' Manual - 6th Edition (2003)
478
Plate girders
17.4.5 Resistance of a web to combined effects
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
If the moment resistance of a plate girder is determined using the high shear general
method (see section 17.4.3.2 (3)) the resistance of the web to combined effects
should be checked in accordance with Annex H.3 of the Code.
The interaction between bending and shear in plate girders is illustrated in Fig.
17.3. The broken line represents the high shear flanges only method (see section
17.4.3.2 (2)) while the full line represents the low shear and high shear general
methods.
flanges only method
Vb
vw
0-5 V
Mf
Fig. 17.3
Interaction between shear and moment resistance of plate girder webs
17.4.6 End panels and end anchorage
17.4.6.1 General
End anchorage need not be provided if either the shear resistance Pv (see section
17.4.4.1) not the shear buckling resistance Vw (see section 17.4.4.2 (1)) is the governing design criterion, indicated by Pv £ Vw, or the applied shear force Fv is less
than the critical shear buckling resistance Vcr (see section 17.4.4.2 (3)). For all other
situations some form of end anchorage is required to resist the post-critical membrane stresses (tension field) which develop in a buckled web.
Three alternatives for providing end anchorage are recommended in Annex H.4
of the Code, namely single stiffener end posts, twin stiffener end posts and anchor
Steel Designers' Manual - 6th Edition (2003)
Design of plate girders used in buildings to BS 5950: Part 1: 2000
—
ten sion
field
flq
479
—
end post
(a)
—
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
—> —
—end post
(b)
-
——
Hq
—-
-- —
anchor
panel
(c)
Fig. 17.4
End anchorage (a) single stiffener end post (b) twin stiffener end post (c) anchor
panel
panels (see Fig. 17.4). End anchorage should be provided for a longitudinal anchor
force Hq representing the longitudinal component of the tension field, at the ends
of webs without intermediate stiffeners and at the end panels of webs with intermediate transverse stiffeners.
If the web is fully loaded in shear (Fv ≥ Vw):
Vcr ˆ
Ê
H q = 0.5dtpyw 1 Ë
Pv ¯
1 2
If the web is not fully loaded in shear:
H q = 0.5dtpyw
Vcr ˆ
Ê Fv - Vcr ˆ Ê
1Ë Vw - Vcr ¯ Ë
Pv ¯
1 2
Steel Designers' Manual - 6th Edition (2003)
480
Plate girders
Each form of end anchorage has to be designed to resist the horizontal component
of the tension field, together with compressive forces due to the support reactions.
17.4.6.2 Single stiffener end post
A single stiffener end post (see Fig. 17.4 (a)) should be designed to resist the support
reaction plus an in-plane bending moment Mtf due to the anchor force Hq given in
general by:
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Mtf = 0.15Hqd
However, if the end post is connected to the girder flange by full strength welds,
and both the width and thickness of the flange are not less than those of the end
post, then a lower value of Mtf given by:
Mtf = 0.1Hqd
may be adopted.
17.4.6.3 Twin stiffener end post
A twin stiffener end post (see Fig. 17.4 (b)) should be designed as a beam spanning
between the flanges of the girder and subjected to a horizontal shear force Rtf given
by:
Rtf = 0.75Hq £ Vcr.ep
where Vcr.ep is the critical shear buckling resistance of the web of the end post.
The end stiffener should be designed to resist the relevant compressive force Fe
due to the support reaction plus a compressive force Ftf due to bending of the end
post given by:
Ftf =
0.15H q d
ae
where ae is the spacing of the two end stiffeners.
The other stiffener forming part of the end post should be designed to resist the
relevant compressive force Fs due to the support-reaction. However, if the tensile
force due to bending of the end post Ftf > Fs then it should be checked for a tensile
force equal to (Ftf - Fs).
17.4.6.4 Anchor panel
An anchor panel (see Fig. 17.4 (c)) should satisfy the condition that the applied
shear force Fv is less than the critical shear buckling resistance Vcr, i.e.
Fv £ Vcr
Steel Designers' Manual - 6th Edition (2003)
Design of plate girders used in buildings to BS 5950: Part 1: 2000
481
In addition the anchor panel should be considered as a beam spanning between the
flanges of the girder and satisfy the condition (see section 17.4.6.3)
Rtf = 0.75Hq £ Vcr.ep
The two transverse stiffeners bounding the anchor panel should be designed in a
similar manner to the stiffeners of a twin end post.
17.4.7 Web stiffeners
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
17.4.7.1 Types of stiffeners
Transverse stiffeners are generally required to ensure the satisfactory performance
of the web panels of slender plate girders. The three most important types of stiffeners are as follows.
(a) Intermediate transverse
(b) Load-carrying
(c) Bearing
Stiffeners of each of these types are subjected to compression and should be
checked for bearing and buckling. A particular stiffener may also serve more than
one function and should therefore be designed for combined effects: e.g. an intermediate transverse stiffener may also be load carrying.
The design of each of the above types of stiffener is detailed in clauses 4.4.6 and
4.5 of the Code.
17.4.7.2 Intermediate transverse web stiffeners
Intermediate transverse web stiffeners are used to increase both the critical shear
buckling strength qcr and post-critical shear buckling strength qw of slender web
panels. They are designed for minimum stiffness and buckling resistance, as specified in clause 4.4.6 of the Code.
Intermediate transverse web stiffeners not subject to external loads should have
a minimum second moment of area Is about the centreline of the web given by:
a
≥ 2
d
a
For < 2
d
For
3
I s = 0.75dt min
2
3
I s = 1.5(d a) dt min
where tmin is the minimum required web thickness for the actual stiffener spacing a.
Steel Designers' Manual - 6th Edition (2003)
482
Plate girders
The buckling resistance of intermediate transverse web stiffeners not subject to
external forces and moments should satisfy the condition
Fq = Vmax - Vcr £ Pq
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
where Fq is the compressive axial force, assumed equal to the larger of the shears
in the two web panels adjacent to the stiffener minus the critical shear buckling
resistance of the same web panel, and Pq is the buckling resistance of the stiffener
from clause 4.5.5 of the Code.
Additional requirements are specified for the minimum stiffness and buckling
resistance of intermediate transverse web stiffeners subject to external forces and
moments.
17.4.7.3 Load-carrying stiffeners
Load-carrying stiffeners are provided to prevent local buckling of the web due to
concentrated loads or reactions applied through the flange. They should be positioned wherever such actions occur, if the resistance of the unstiffened web would
otherwise be exceeded.
Load-carrying stiffeners must be checked for both bearing and buckling, as
specified in clauses 4.5.2 and 4.5.3 of the Code. When checking for buckling an effective web width of 15t on either side of the centreline of the stiffener is considered
to act with it to form a cruciform section.The resulting stiffener strut is then assumed
to have an effective length of 0.7 times its actual length if the loaded flange is
restrained against rotation in the plane of the stiffener. If the loaded flange is not
restrained the effective length is taken equal to the actual length. The buckling
resistance of the stiffener strut is then determined as for a normal compression
member.
When checking the stiffener for bearing, only that area of the stiffener in contact
with the flange should be taken into account.
Sometimes it is not possible to provide stiffeners immediately under the externally applied loading, e.g. a travelling load on a gantry girder or loads applied during
launching. In such cases of patch loading (loads applied between the stiffeners) an
additional check is required to ensure that the specified compressive strength of the
web for edge loading is not exceeded.
17.4.7.4 Bearing stiffeners
Bearing stiffeners are provided to prevent local crushing of the web due to concentrated loads or reactions applied through the flange. If such actions do not exceed
the buckling resistance of the web, then the load-carrying stiffeners of section
17.4.7.3 are not needed. However, the local bearing resistance may still be exceeded.
Steel Designers' Manual - 6th Edition (2003)
Initial choice of cross-section for plate girders used in bridges
483
If so then bearing stiffeners should be provided and designed in accordance with
clause 4.5.2 of the Code.
17.4.8 Gantry girders
Plate girders used to support cranes should be designed in accordance with clauses
4.4 and 4.11 of the Code.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
17.5 Initial choice of cross-section for plate girders used in bridges
17.5.1 Choice of span
Plate girders are frequently employed to support railway and highway loadings on
account of their economic advantages and ease of fabrication, and are considered
suitable for spans in the region of 25–100 m. Plate girders are considered when rolled
sections are not big enough to carry the loads over the chosen span.
Spans are usually fixed by site restrictions and clearances. If there is freedom for
the designer, simply-supported spans within the range of 25–45 m will be found to
be appropriate; the optimum for continuous spans is about 45 m, as 27 m long girders
can be spliced with pier girders 18 m long. However, plate girders can be employed
for spans of up to about 100 m in continuous construction.
17.5.2 Span-to-depth ratios
For supporting highways a composite concrete decking, having a concrete thickness
in the region of 250 mm, is commonly employed. A span-to-depth ratio of 20 for
simply-supported spans serves as an initial choice for such girders; where continuous spans are employed the depth of the girder can be reduced at least by one-third
compared with a simply-supported span.
Through girders of constant depth are rarely employed in continuous construction. For simply supported plate girders designed for highway loading, a span-todepth ratio of 20 can serve as an initial choice; for through girders supporting railway
loading larger depths are required, and a span-to-depth ratio of 15 is more appropriate. Variable-depth plate girders are more appropriate where spans in excess of
30 m are required.
For continuous girders in composite construction, a span-to-depth ratio of 25
at the pier and 40 at mid-span is suitable. For highway bridges provided with
an orthotropic deck, the corresponding values are increased to 30 and 60
respectively.
Steel Designers' Manual - 6th Edition (2003)
484
Plate girders
Table 17.2 Guide for selection of deck thickness
Transverse girder
spacing (m)
2.5–3.8
3.3–5.5
5.0–7.0
6.5+
Deck
Reinforced concrete slab on
permanent formwork
Haunched slab
Stringer
Cross girders
Thickness (mm)
Constant 225–275
Min. 250 up to 350 at haunch
Constant 225–275
Typically 250
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
17.5.3 Initial sizing of the flanges
It is important to keep the flanges as wide as possible consistent with outstand
limitations. The oustand should not ordinarily be wider than 12 times its thickness
if the flange is fully stressed in compression, or 16 times its thickness if the flange is
not fully stressed or is in tension. This ensures that the flange is stable during erection with the minimum number of bracings. For practical reasons, e.g. to accommodate detailing for certain types of permanent formwork, it is desirable to use a
minimum flange width of 400 mm. A maximum flange thickness of 65 mm is recommended to avoid heavy welds and the consequent distortion.
Changes in the width of the top flange can be incorporated easily in composite
decks to suit design requirements, and these do not invite criticism on appearance
grounds. On the other hand, changes in widths of bottom flanges are less acceptable visually. In any case it is desirable not to change flange sizes frequently lest
the economy achieved by saving in material should be offset by expensive butt
welding.
For composite girders having concrete decks it is usually necessary to allow for
at least two rows of shear connectors on top flanges of beams; for longer spans three
rows may be required at piers where high shear transfer takes place.
The cross-sections chosen affect deck thickness and overall structural form. Table
17.2 provides useful guidance for the initial selection of deck thickness.
17.5.4 Initial sizing of the web
The initial choice of web thickness is related to stiffening. There is no special advantage in using too slender a web, as the material saved will have to be replaced by
stiffening; moreover, the workmanship with a thicker web plate is often superior.
Probably the biggest single problem when determining the girder layout for a concrete decked plate girder road bridge is to achieve a solution with optimum transverse deck spans and minimum cost of web steel, as this involves the balancing of
weight against workmanship.
The following advice is intended as general guidance.
Steel Designers' Manual - 6th Edition (2003)
Design of steel bridges to BS 5400: Part 3
485
Table 17.3 Initial values of web thickness
Beam depth (mm)
Up to 1200
1200–1800
1800–2250
2250–3000
Web thickness (mm)
10
12
15
20
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Consideration should also be given to the economy
of providing thicker webs, without any intermediate
stiffening.
(a) Two or three vertical stiffeners should be provided, at a spacing of approximately one times the girder depth, close to the bearings; thereafter their spacing
should be increased to 1.5 times the girder depth.
(b) In general horizontal stiffeners should be avoided. In long span continuous plate
girders they may be necessary in locations close to the piers.
(c) It is desirable to provide vertical stiffeners on one side and horizontal stiffeners on the other side, where possible.
(d) Initial values for web thickness are suggested in Table 17.3, if it is intended to
provide web stiffening. Consideration should also be given to the economy of
providing thicker webs, without any intermediate stiffening.
(e) Compact girders are more economical for most simply supported spans and for
shorter continuous spans; economical plate girders for longer continuous spans
will be non-compact.
17.6 Design of steel bridges to BS 5400: Part 3
17.6.1 Global analysis
The Code requires that the global analysis of the structure should be carried out
elastically to determine the load effects, i.e. bending moments, shear forces etc. The
section properties to be used will generally be those of the gross section (see clauses
7.1 and 7.2 of the Code2).
Plastic analysis of the structure, i.e. redistribution of moments due to plastic hinge
formation, is not allowed under BS 5400: Part 3.
Analysis should be carried out for individual and unfactored load cases. Summation of the load effects in different combinations and with different load factors can
then be carried out in tabular form, as required by the design process.
17.6.2 Design of beams at the ultimate limit state
17.6.2.1 Basis of design
In the design of beam cross sections at the ultimate limit state the following need
to be considered.
Steel Designers' Manual - 6th Edition (2003)
486
Plate girders
(a) Material strength
(b) Limitations on shape on account of local buckling of individual elements (webs
and flanges)
(c) Effective section (reductions for compression buckling and holes)
(d) Lateral torsional buckling
(e) Web buckling (governed by depth-to-thickness ratio of web and panel aspect
ratio)
(f) Combined effects of bending and shear.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
17.6.2.2 Material strength (clause 4.3 of the Code)
The nominal material strength is the yield stress of steel. The partial factor on
strength gm depends on the structural component and behaviour; gernerally it is 1.05
at the ultimate limit state, but higher values are specified in the Code for compressive stress in bending or buckling.
17.6.2.3 Shape limitations (clause 9.3 of the Code)
The resistance of a section can be limited by local buckling if the flange outstand to
thickness ratio is large.The Code limits this ratio so that local buckling will not govern.
Similarly, limits are given for outstands in tension to limit local shear lag effects.
Where the compression region of a web without stiffeners has a very large depth
to thickness ratio, i.e. > 68, the bending resistance of the section is modified by the
requirement that webs are effectively reduced to take account of local buckling.
Webs with longitudinal stiffeners are not so reduced.
When calculating the bending resistance of beams, it is important to recognize
the contribution made by the webs and make appropriate provision for it in
computing the modulus of the ‘effective section’. Slender webs without horizontal
stiffeners buckle in the compression zone when subjected to high bending
stresses. Based on parametric studies, the bending resistance contributed by the web
(associated with a flange) is assessed by reducing the thickness of the web using the
formulae
t we
=1
tw
if
yc Ê s yw ˆ
Á
˜
t w Ë 355 ¯
1 2
£ 68
1 2
1 2
t we È
yc Ê s yw ˆ ˘
yc Ê s yw ˆ
= Í1.425 - 0.00625 Á
˜ ˙ if 68 < Á
˜ < 228
t w ÍÎ
t w Ë 355 ¯ ˙˚
t w Ë 355 ¯
t we
=0
tw
if
yc
tw
Ê s yw ˆ
Á
˜
Ë 355 ¯
1 2
≥ 228
where yc is the depth of the web measured in its plane from the elastic neutral axis
of the gross section of the beam to the compressive edge of the web.
Steel Designers' Manual - 6th Edition (2003)
Design of steel bridges to BS 5400: Part 3
487
b07 tf0( 355/
NA
Til
dna 28tw(355/f)°5
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
- _tw
Fig. 17.5
Section geometry and classification – compact sections
The Code further defines cross-sections as compact or non-compact, and a different design approach is required for each classificatiod.7
The classification of a section depends on the width to thickness ratio of the elements of the cross-section considered, as shown in Fig. 17.5 for a compact plate
girder.
A compact cross-section can develop the full plastic moment resistance of the
section, i.e. a rectangular stress block, and local buckling of the individual elements
of the cross section will not occur before this stage is reached (see Fig. 17.6 (b)).
However, in a non-compact section local buckling of elements of the cross section
may occur before the full moment resistance is reached, and hence the design of
such sections is limited to first yield in the extreme fibre i.e. a triangular stress block
(see Fig. 17.6 (c)).
These classifications can best be illustrated by considering a beam in fully
restrained bending. This gives the idealized moment resistance of the cross sections
as follows.
Compact sections
M1 =
Non-compact sections M1 =
Zpes y
g m g f3
Zs y
g m g f3
where sy is the yield stress, Zpe is the plastic section modulus, Z is the elastic section
modulus, gm is the partial factor on strength, and gf3 is the partial factor on loads,
reflecting the uncertainty of loads.
The use of the plastic section modulus for compact sections does not imply that
plastic analysis can be employed; in fact it is specifically excluded by BS 5400: Part
3. The achievement of a rectangular stress block does not necessarily mean that
there has been redistribution of moments along the member.
Steel Designers' Manual - 6th Edition (2003)
488
Plate girders
yi
(b)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 17.6
(c)
Design stresses for compact and non-compact sections (a) beam cross-section
(b) compact section (c) non-compact section
17.6.2.4 Lateral torsional buckling (clause 9.6 of the Code)
Where a member has portions of its length with unrestrained elements in compression, lateral torsional buckling may occur (see Fig. 17.7). The Code deals with
these effects by the use of a slenderness parameter lLT defined by
l LT =
Le
K4hn
ry
where Le is the effective length for lateral torsional buckling, ry is the radius of gyration of the whole beam about its minor y-y axis, K4 is a torsion factor (taken as 0.9
for rolled I or channel sections and 1.0 for all other sections), h is a factor that takes
account of moment gradient, i.e. the shape of the moment diagram and the fact that
a uniform moment over the unrestrained length will cause buckling more readily
than a non-uniform moment, and n is a torsion factor dependent on the shape of
the beam.
The process of calculating the value of the limiting compressive stress slc
corresponding to the value of lLT is described in the Code.
Portions of beams between restraints can deflect downwards, sideways and rotate.
Failure may then occur before the full moment resistance of the section is reached.
The possibility of this type of failure is dictated by the unrestrained length of
the compression flange, the cross-section geometry of the beam and the moment
gradient.
17.6.2.5 Shear resistance (clause 9.9.2 of the Code)
The ultimate shear resistance VD of a web panel under pure shear should be
taken as
Steel Designers' Manual - 6th Edition (2003)
Design of steel bridges to BS 5400: Part 3
489
Lateral restraint
support
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
quarter length of buckled beam
-section at support
Fig. 17.7
1w
section at restraint
section between restraints
(vertical displacement)
(vertical horizontal and
rotational displacements)
Lateral torsional buckling
È t w (dw - hh ) ˘
VD = Í
tl
Î g m g f3 ˙˚
where tw and dw are the thickness and depth of the web respectively, hh is the height
of the largest cut-out within the panel being considered, gm and gf3 are partial material and load factors respectively, and tl is the limiting shear stress. For slender
webs tl is governed by shear buckling, which becomes significant in webs with depth
to thickness ratios greater than about 80.
The actual shear strength of a web panel is dependent on the following:
(a)
(b)
(c)
(d)
Yield stress (suitably factored)
Depth-to-thickness ratio of web
Spacing of stiffeners
Conditions of restraint provided by flanges
Steel Designers' Manual - 6th Edition (2003)
490
Plate girders
tension field stress
plastic hinge
\
yield zone
(b)
(a)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fig. 17.8
(c)
Tension field theory (a) pure shear (b) tension field (c) collapse mechanism
The web of a plate girder between stiffeners acts similarly to the diagonal of
a Pratt truss (this phenomenon, known as tension field action, is described in
Reference 6). The theory stipulates that the web will resist the applied loading
in three successively occurring stages (see Fig. 17.8):
Stage 1 A pure shear field
Stage 2 A diagonal tension field
Stage 3 A collapse mechanism due to the formation of plastic hinges in the flanges
In slender web panels the limit of stage 1, i.e. the pure shear field, is reached when
the applied shear stress reaches the elastic critical stress tc. To allow for the effects
of residual stresses and initial imperfections in stocky plates, tc is limited to less than
its actual value when it is greater than 0.8 times the shear yield stress ty, and to ty
when its actual value exceeds 1.5 times ty.
The elastic critical stress tc for a plate loaded in shear is given by
tc = K
p 2E Ê tw ˆ
12(1 - n 2 ) Ë dwe ¯
2
where E and n are the elastic modulus and Poisson’s ratio of the steel respectively,
and
K = 5.34 +
K =4+
4
f2
5.34
f2
when f ≥ 1
when f < 1
f = panel aspect ratio =
a
dwe
The criteria outlined previously have been incorporated in computing the value of
tc as follows:
Steel Designers' Manual - 6th Edition (2003)
Design of steel bridges to BS 5400: Part 3
t c 904
= 2
ty
b
when b ≥ 33.62
=1
when b £ 24.55
= 1.54 - 0.022b
when 24.55 < b < 33.62
491
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
where
b=
l
K1 2
l=
dwe Ê s yw ˆ
Á
˜
t w Ë 355 ¯
1 2
In stage 2 a tensile membrane stress field develops in the panel, the direction of
which does not necessarily coincide with the diagonal of the panel. The maximum
shear resistance is reached in stage 3, when the pure shear stress of stage 1 and the
membrane stress of stage 2 cause yielding of the panel according to the von Mises
yield criterion, and plastic hinges are formed in the flanges.
The magnitude of the membrane tensile stress st in terms of its assumed direction q and the first stage limiting stress tc is given by
2
st È
Ê tc ˆ ˘
= Í3 + (2.25 sin 2 2q - 3)Á ˜ ˙
Ëty ¯ ˚
ty Î
1 2
- 1.5
tc
sin 2q
ty
Adding the resistance in the three stages, the ultimate shear strength tu is obtained
as
1 2
˘
t u Èt c
st ˆ
st
Ê
= Í + 5.264 sin q Á mfw ˜ + (cot q - f ) sin 2 q ˙
Ë
t y Ît y
ty ¯
ty
˚
when
mfw £
f2 st
sin 2 q
4 3 ty
and
t u È 4 3 mfw s t
tc ˘
=Í
+
sin 2 q + ˙
ty Î f
2t y
ty ˚
when
mfw >
f2 st
sin 2 q
4 3 ty
To provide an added measure of safety in respect of slender webs, the above values
are multiplied by a varying correction factor f to obtain the limiting shear stress.
Steel Designers' Manual - 6th Edition (2003)
Plate girders
492
f =1
1.15
f=
1.35
f=
when l £ 56
when l ≥ 156
1.15
1.15 + 0.002(l - 56)
when 56 < l < 156
The value of tl/ty is taken as the lower of tu/ty computed as above and 1.0.
The term mfw in the above equations is a non-dimensional representation of the
plastic moment resistance of the flange (taking the smaller value of the top and
bottom flanges, ignoring any concrete).
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
mfw =
=
Mp
d t s yw
2
we w
s yf bfe ff2
2
2dwe
t ws yw
Values of tl/ty versus l are plotted in Figures 11–17 of the Code corresponding to
various values of mfw and f. These can be used directly by designers.
The term bfe is the width of flange associated with the web and is limited to
Ê 355 ˆ
bfe £ 10t f Á
˜
Ë s yf ¯
1 2
Where two flanges are unequal, the value of mfw is taken conservatively as the value
corresponding to the weaker flange. Moreover, only a section symmetric about the
mid-plane is taken as effective; any portion of the flange plate outside this plane
of symmetry is ignored so that complexities due to the torsion of the flange are
eliminated.
The above procedure has to be repeated for several values of q, and the highest
value of tu/ty is to be used. From parametric studies it has been established that q
is never less than (1/3)cot-1 f or more than (4/3)cot-1 f.
When tension field action is used, consideration must be given to the anchorage
of the tension field forces in the end panels, and special procedures must be adopted
for designing the end stiffener.
17.6.2.6 Combined bending and shear (clause 9.9.3 of the Code)
The Code has simplified the procedure for girders without longitudinal stiffeners
by allowing shear and bending resistance to be calculated independently and then
combined by employing an interaction relationship on the basis given below.
(a) The bending resistance of the whole section is determined with and without
contribution from the web (Ml and MR respectively).
Steel Designers' Manual - 6th Edition (2003)
Design of steel bridges to BS 5400: Part 3
493
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
(b) The shear resistance using the tension field theory discussed above is determined
with and without contribution from the flanges (VD and VR respectively).
(c) The bending and shear resistance without any contribution from the web and
flanges respectively (MR and VR) can be mobilized simultaneously.
(d) The pure bending resistance of the whole section MD can be obtained, even
when there is a coincident shear on the section, provided the latter is less than
1
/2VR. Similarly the theoretical design shear resistance VD of the whole beam
section is attained provided the coexisting bending moment is not greater than
1
/2MR.
The interaction relationship is linear between this set of values of shear force and
bending moment and is shown graphically in Fig. 17.9. The important feature from
the designer’s point of view is that full values of calculated bending resistance can
be utilized in the presence of a moderate magnitude of shear.
VD
VI2
/2
Fig. 17.9
Interaction between shear and bending resistance
For girders with longitudinal stiffeners, account is taken of the combined effects
of bending and shear by comparison of the stresses in the web panels with the
relevant buckling strength of the panel. The buckling coefficients are based upon
large-deflection elastic–plastic finite element studies. An interaction expression
is then used which is based on an equivalent stress check. This makes suitable
allowance for webs with longitudinal stiffeners and having large depth-to-thickness
ratios; no reduction need be made to account for local buckling.
17.6.2.7 Bearing stiffeners (clause 9.14 of the Code)
BS 5400: Part 3 requires such stiffeners to be provided at supports. The design forces
to be applied to the stiffener include the direct reaction from the bearing (less the
Steel Designers' Manual - 6th Edition (2003)
494
Plate girders
force component in the flange if the beam soffit is haunched), a force to account
for the destabilizing effect of the web, forces from any local cross bracing (including forces from restraint systems), and a force due to the tension field. The derivation of these is far more precise and lengthy than BS 5950: Part 1: 2000 requires,
largely because plate girders in bridges are usually part of a much more complex
structural system than in buildings.
Bearing stiffeners are almost always placed symmetrically about the web centreline. Stiffeners that are symmetrical about the axis perpendicular to the web, i.e. flats
or tees rather than angles, are preferred.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
17.6.2.8 Intermediate stiffeners (clause 9.13 of the Code)
Intermediate stiffeners are usually placed on one side of the web only. Standard flat
sizes should be used. BS 5400: Part 3 does not prescribe stiffness criteria for intermediate stiffeners (which have an associated length of web), but it does prescribe
design forces for which these stiffeners are designed, similar to those outlined for
bearing stiffeners.
17.6.3 Design of beams at the serviceability limit state
In practice, due to the proportions of partial safety factors at the ultimate and
serviceability limits, the serviceability check on the steel section is automatically
satisfied if designs are satisfactory at the ultimate stage.
The design of beams is carried out using an elastic strain distribution throughout
the cross-section, in a manner generally similar to that used at the ultimate limit but
with different partial safety factors, i.e. different values of gfL, gm and gf3. The determination of effective section must take account of shear lag, where appropriate.
Stresses should be checked at critical points in the steel member to ensure
that no permanent deformation due to yielding takes place.8–10 Plastic moment
resistance is not considered at the serviceability limit state.
In composite construction, crack widths in the concrete deck may often govern
the design at the serviceability limit state. For non-compact sections the analysis at
the ultimate limit state is carried out using a triangular (elastic) stress distribution.
Consequently the serviceability limit check for bending is always satisfied, since the
factors are less than those at the ultimate limit state. The Code lists the clauses which
require checking at both limit states.
17.6.4 Fatigue
Fatigue should be checked with reference to BS 5400: Part 10. Generally it is only
bracing connections, stiffeners and shear connectors, and their welding to the
Steel Designers' Manual - 6th Edition (2003)
References
495
girders, that have to be checked for fatigue. The details of the connection may have
a significant effect on the fatigue life, and by careful detailing fatigue may be
‘designed out’ of the bridge.
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
17.6.5 Design format
One of the significant features of BS 5400: Part 3 is that plastic redistribution across
the section is now permitted, in certain circumstances, prior to the attainment of
the ultimate limit state. Such redistribution is much less than permitted in simple
plastic design of buildings, but can give a considerable enhancement in strength. For
example, the design bending strength of compact sections is based on the plastic
section modulus, and the basic shear strength is based on rules which take account
of tension field action, which only develops in the presence of considerable plasticity. The interaction between shear strength and bending strength of a cross-section
is empirically based and implicitly assumes that plastic strains may occur prior to
the attainment of the design strength. Finally, the design strength of compact sections that are built in stages assumes that a redistribution of stresses may take place
within the cross-section. The ultimate limit state check for the completed structure
is simply to ensure that the bending resistance (given by the limiting stress times
the plastic section modulus divided by partial factors) is greater than the maximum
design moment (obtained by summing all the moments due to the various design
loads).
This recognition of the reserve of strength beyond first yield is limited to situations
where plasticity may be permitted safely. The principal limitations are as follows.
(a) Plastic section moduli may only be used for compact sections, i.e. those that can
sustain local compressive yielding without any local buckling.
(b) Where the structure is built in stages and is non-compact, plasticity is permitted implicitly only in considering the interaction between bending and shear
strength.
(c) Where longitudinal stiffeners are present, neither the plastic section modulus
nor the plastic bending / shear interaction may be used.
(d) If the flanges are curved in elevation, similar restrictions to (c) apply.
References to Chapter 17
1. British Standards Institution (2000) BS 5950: Structural use of steelwork in
building: Part 1: Code of practice for design: rolled and welded sections. BSI,
London.
2. British Standards Institution (1982) BS 5400: Steel concrete and composite
bridges: Part 3: Code of practice for the design of steel bridges. BSI, London.
Steel Designers' Manual - 6th Edition (2003)
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
496
Plate girders
3. Owens G.W. (1989) Design of Fabricated Composite Beams in Buildings. The
Steel Construction Institute, Ascot, Berks.
4. Narayanan R., Lawless V., Naji F.J. & Taylor J.C. (1993) Introduction to Concise
Eurocode 3 (C-EC3) with worked examples. SCI Publication 115, The Steel
Construction Institute, Ascot, Berks.
5. Lawson R.M. & Rackham J.W. (1989) Design for Openings in Webs of Composite Beams. The Steel Construction Institute, Ascot, Berks.
6. Evans H.R. (1988) Design of plate girders. In Introduction to Steelwork Design
to BS 5950: Part 1, pp. 12.1–12.10. The Steel Construction Institute, Ascot, Berks.
7. Chaterjee S. (1981) Design of webs and stiffeners in plate and box girders. In
The Design of Steel Bridges (Ed. by K.C. Rockey & H.R. Evans), Chapter 11.
Granada.
8. Iles D.C. (1989) Design Guide for Continuous Composite Bridges 1: Compact
Sections. SCI Publication 065, The Steel Construction Institute, Ascot, Berks.
9. Iles D.C. (1989) Design Guide for Continuous Composite Bridges 2: NonCompact Sections. SCI Publication 066, The Steel Construction Institute, Ascot,
Berks.
10. Iles D.C. (1991) Design Guide for Simply Supported Composite Bridges. SCI
Publication 084, The Steel Construction Institute, Ascot, Berks.
Further reading for Chapter 17
Iles D.C. (2001) Design of Composite Bridges: General Guidance. SCI Publication
289, The Steel Construction Institute, Ascot, Berks.
Iles D.C. (2001) Design of Composite Bridges: Worked Examples. SCI Publication
290, The Steel Construction Institute, Ascot, Berks.
A worked example follows which is relevant to Chapter 17.
Steel Designers' Manual - 6th Edition (2003)
Worked examples
Subject
The
Steel Construction
Institute
Chapter ref.
PLATE GIRDERS
17
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
Made by TMR
Checked by
Sheet no.
1
BD
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
WE 17.1 Design brief
The girder shown in Fig.WE 17.1 is fully restrained throughout
its length. For the loading shown and specified in WE 17.2 design
a transversely stiffened plate girder in grade S275 steel. Girder
depth is unrestricted.
0350
11300
H
30 000
Fig.WE 17.1
w
Plate girder span and loading
= wd + wi
W1 = W1d + W1i
W2 = W2d + W2i
WE 17.2 Loading
Dead loads
Imposed loads
UDL
wd = 20 kN/m
Point loads
W1d = 200 kN
W2d = 200 kN
UDL
wi = 40 kN/m
Point loads
W1i = 400 kN
W2i = 400 kN
2.4.1.1
Table 2
WE 17.3 Load factors
Dead load factor
gfd = 1.4
Imposed load factor
gfi = 1.6
497
Steel Designers' Manual - 6th Edition (2003)
498
Worked examples
Subject
The
Steel Construction
Institute
Chapter ref.
PLATE GIRDERS
17
Silwood Park, Ascot, Berks SL5 7QN
Made by TMR
Design code
BS 5950: Part 1
Checked by
WE 17.4 Factored loads
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
w = wdgfd + wigfi
W¢1 = W1dgfd + W1igfi = 200 ¥ 1.4 + 400 ¥ 1.6
= 920 kN
W¢2 = W2dgfd + W2igfi = 200 ¥ 1.4 + 400 ¥ 1.6
= 920 kN
WE 17.5 Design shear forces and moments
The design shear forces and moments corresponding to the factored
loads are shown in Fig.WE 17.2.
92kN/m
92QkN
/'
//"
I
121' 3900 3900
I
I
5050
30000
AE I
Shear force kN
18952
Bending moment kNm
Fig.WE 17.2
2
2.4.1.1
Table 2
= 20 ¥ 1.4 + 40 ¥ 1.6
= 92 kN/m
92OkNI
Sheet no.
BD
Design shear forces, moments and stiffener spacing
Steel Designers' Manual - 6th Edition (2003)
Worked examples
Subject
The
Steel Construction
Institute
Chapter ref.
PLATE GIRDERS
17
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
Made by TMR
Checked by
Sheet no.
3
BD
WE 17.6 Initial sizing of plate girder
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
The recommended span/depth ratio for simply supported noncomposite girders varies between 12 for short span girders and 20
for long span girders. Herein the depth is assumed to be span/15.
d=
span 30 000
=
= 2000 mm
15
15
Estimate flange area assuming pyf = 255 N/mm2
Af =
Mmax 18952 ¥ 10 6
=
= 37160 mm2
dpyf
2000 ¥ 255
3.1.1
Table 9
4.4.4.2(b)
For non-composite girders the flange width is usually within the
range 0.3 to 0.5 of the depth. Assume a flange 750 ¥ 50 = 37500 mm2.
The minimum web thickness for plate girders in buildings usually
satisfies t ≥ d/120. Assume the web thickness t = 15 mm, slightly
less than d/120 due to the proposed transverse stiffening.
WE 17.7 Section classification
Assume girder with flanges 750 ¥ 50 mm and web 2000 ¥ 15 mm
Flange
For T = 50 mm
Ê 275 ˆ
e=Á
˜
Ë pyf ¯
12
pyf = 255 N/mm2
275 ˆ
=Ê
Ë 255 ¯
Table 9
12
= 1.04
Table 11 Note 2
750 - 15
= 367.5 mm
2
b 367.5
=
= 7.35
T
50
b=
Compact limiting value of b/T for welded sections is 9e = 9 ¥ 1.04
= 9.36.
Table 11
499
Steel Designers' Manual - 6th Edition (2003)
500
Worked examples
Subject
The
Steel Construction
Institute
Chapter ref.
PLATE GIRDERS
17
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
Made by TMR
Checked by
Sheet no.
4
BD
b
= 7.35 < 9.36
T
\ Flange is compact
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Web
For t = 15 mm
pyw = 275 N/mm2
e=1
Table 9
d 2000
=
= 133
t
15
d
= 133 > 120 e = 120
t
\ Web is slender
Since
Table 11
d
> 63e the web must be checked for shear buckling
t
WE 17.8 Dimensions of web and flanges
4.4.5.1
4.4.3
Assume a stiffener spacing a > 1.5d
Minimum web thickness to avoid serviceability problems
t = 15 ≥
d
2000
=
= 8.0 mm
250
250
4.4.3.2(b)
OK
To avoid the flanges buckling into the web
4.4.3.3(b)
d ˆ Ê p yf ˆ
t = 15 ≥ Ê
Ë 250 ¯ Ë 345 ¯
2000 ˆ Ê 255 ˆ
= 5.91 mm
=Ê
Ë 250 ¯ Ë 345 ¯
4.4.4
WE 17.9 Moment resistance
For sections with slender webs (and flanges which are not slender)
three methods of design are specified. Method (b) is used herein.
The moment is assumed to be resisted by the flanges alone while
the web is designed for shear only.
Mf = pyfAfhs
where hs = 2000 + 50 = 2050 is the distance between the centroids
of flanges
Mf = 255 ¥ 750 ¥ 50 ¥ 2050 ¥ 10-6 = 19603 kNm
Mf = 19603 > Mmax = 18952 kNm
OK
4.4.4.2(b)
Steel Designers' Manual - 6th Edition (2003)
Worked examples
Subject
The
Steel Construction
Institute
Chapter ref.
PLATE GIRDERS
17
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
Made by TMR
Checked by
Sheet no.
5
BD
WE 17.10 Shear buckling resistance of web
4.4.5
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Webs with intermediate transverse stiffeners may be designed
according to 4.4.5.2 or 4.4.5.3. The simplified method of 4.4.5.2 is
used herein. The assumed stiffener spacing is shown in Fig.WE 17.2.
WE 17.11 Shear buckling resistance of end panel AB
The Code allows three alternative methods for providing end
anchorage. The most commonly employed method is to design end
panels without accounting for tension field action. For end panel
AB:
d = 2000 mm
t = 15 mm
a 1550
=
= 0.775
d 2000
Fv = 2300 kN
d 2000
=
= 133
t
15
pyw = 275 N/mm2
Pv = 0.6pywd t = 0.6 ¥ 275 ¥ 2000 ¥ 15 ¥ 10-3 = 4950 kN
qw = 132 N/mm2
4.2.3
Table 21
Vw = qwd t = 132 ¥ 2000 ¥ 15 ¥ 10-3 = 3960 kN
4.4.5.2
Vw 3960
=
= 0.8
4950
Pv
Vcr =
9Vw - 2 Pv 9 ¥ 3960 - 2 ¥ 4950
=
= 3677 kN
7
7
Fv = 2300 < Vcr = 3677 kN
OK
Fv = 2300 < Pv = 4950 kN
OK
End panel AB should also be checked as a beam spanning between
the flanges of the girder capable of resisting a shear force Rtf due
to the anchor forces in panel BC.
4.4.5.4
H.4.4
Fig. 17.4
501
Steel Designers' Manual - 6th Edition (2003)
502
Worked examples
Subject
The
Steel Construction
Institute
Chapter ref.
PLATE GIRDERS
17
Silwood Park, Ascot, Berks SL5 7QN
Made by TMR
Design code
BS 5950: Part 1
Checked by
Sheet no.
6
BD
Rtf = 0.75 Hq ≥ Vcr.ep
For panel BC Hq is given conservatively by
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Hq = 0.5 d t pyw (1 - Vcr Pv )
H.4.1
12
Calculations presented in WE 17.12 give
Vcr = 1983 kN
1983 ˆ
Hq = 0.5 ¥ 2000 ¥ 15 ¥ 275 ¥ 10 -3 Ê 1 Ë
4950 ¯
12
= 3193 kN
For panel AB spanning between the flanges
a = 2000 mm
d = 1550 mm
t = 15 mm
Pv = 0.6 ¥ 275 ¥ 1550 ¥ 15 ¥ 10-3 = 3836 kN
a 2000
=
= 1.29
d 1550
4.2.3
d 1550
=
= 103
t
15
qw = 132 N/mm2
Table 21
Vw = 1550 ¥ 15 ¥ 132 ¥ 10-3 = 3069 kN
4.4.5.2
Vw 3069
=
= 0.8
3836
Pv
Vcr .ep =
9 ¥ 3069 - 2 ¥ 3836
= 2850 kN
7
4.4.5.4
Check
Rtf = 0.75 ¥ 3193 = 2395 < Vcr.ep = 2850 kN
OK
Steel Designers' Manual - 6th Edition (2003)
Worked examples
Subject
The
Steel Construction
Institute
Chapter ref.
PLATE GIRDERS
17
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
Made by TMR
Checked by
Sheet no.
7
BD
WE 17.12 Shear buckling resistance of panels BC and DE
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
The shear buckling resistance of panel BC is determined using the
simplified method.
d = 2000 mm
t = 15 mm
a 3900
=
= 1.95
d 2000
a = 3900 mm
d 2000
=
= 133
t
15
qw = 94 N/mm2
Vw = d t qw = 2000 ¥ 15 ¥ 94 ¥ 10-3 = 2820 kN
Vw 2820
=
= 0.57
4950
Pv
2
2
Ê Vw ˆ
Ê 2820 ˆ
Ë 0.9 ¯
Ë
¯
Vcr =
= 0.9
= 1983 kN
Pv
4950
Fv = 2206 < Vw = 2820 kN
OK
For panel DE
a 5650
=
= 2.825
d 2000
d 2000
=
= 133
t
15
qw = 89 N/mm2
Vw = 2000 ¥ 15 ¥ 89 ¥ 10-3 = 2670 kN
2
Ê 2670 ˆ
Ë
¯
Vcr = 0.9
= 1778 kN
4950
Fv = 520 < Vw = 2670 kN
OK
4.4.5.4
503
Steel Designers' Manual - 6th Edition (2003)
504
Worked examples
Subject
The
Steel Construction
Institute
Chapter ref.
PLATE GIRDERS
17
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
Made by TMR
Checked by
Sheet no.
8
BD
WE 17.13 Load-carrying stiffener at A
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
The stiffener should be designed for a compressive force Fe due to
the support reaction plus a compressive force Ftf due to the anchor
force
Fx = Fe + Ftf = Fe + 0.15 Hq
Fe = 2300 kN
d
ae
H q = 3193 kN
F x = 2300 + 0.15 ¥ 3193 ¥
H.4.4
( see WE 17.11)
2000
= 2918 kN
1550
Try stiffener consisting of two flats 280 ¥ 22 mm.
Check oustands
4.5.1.2
For ts = 22 < 40 mm pys = 265 N/mm2
Table 9
Ê 275 ˆ
e=Á
˜
Ë pys ¯
12
275 ˆ
=Ê
Ë 265 ¯
12
= 1.02
bs = 280 £ 13 e ts = 13 ¥ 1.02 ¥ 22 = 292 mm
Since bs < 13 e ts design should be based on the net cross-section
area As,net. Allowing 15 mm to cope for the web flange weld:
As ,net = ( 280 - 15 ) 22 ¥ 2 = 11660 mm2
Check bearing resistance of stiffener, neglecting resistance of web
Ps = pysAs,net = 265 ¥ 11660 ¥ 10-3 = 3090 kN
4.5.2.2
Steel Designers' Manual - 6th Edition (2003)
Worked examples
Subject
The
Steel Construction
Institute
Chapter ref.
PLATE GIRDERS
17
Silwood Park, Ascot, Berks SL5 7QN
Made by TMR
Design code
BS 5950: Part 1
Fx = 2918 < Ps = 3090 kN
Checked by
Sheet no.
9
BD
OK
4.5.3.3
Check buckling resistance of stiffener
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Px = As pc
The effective area As includes the area of the stiffeners plus an effective width of web equal to 15t = 15 ¥ 15 = 225 mm (see Fig.WE 17.3).
y
Fig.WE 17.3
y
Load-carrying stiffener at A
As = 2 ¥ 280 ¥ 22 + ( 225 + 11) ¥ 15 = 15860 mm2
Assume the flange of the girder to be restrained against rotation in
the plane of the stiffener.
LE = 0.7d = 0.7 ¥ 2000 = 1400 mm
Determine Iy for stiffener neglecting small contribution from the
web
Iy =
22 ¥ 575 3
= 348 ¥ 10 6 mm 4
12
Determine compressive strength pc .
4.7.5
505
Steel Designers' Manual - 6th Edition (2003)
Worked examples
506
Subject
The
Steel Construction
Institute
Chapter ref.
PLATE GIRDERS
17
Silwood Park, Ascot, Berks SL5 7QN
Made by TMR
Design code
BS 5950: Part 1
Checked by
Since the stiffeners themselves are not welded sections, the reduction of pys by 20 N/mm2 (see 4.7.5) should not be applied
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
l=
Sheet no.
10
BD
4.5.3.3
LE
1400
= 9.45
12 =
6 12
I
Ê yˆ
Ê 348 ¥ 10 ˆ
Ë As ¯
Ë 15860 ¯
pc = 265 N/mm2
Table 24
Px = As pc = 15860 ¥ 265 ¥ 10-3 = 4203 kN
Fx = 2918 < Px = 4203 kN
OK
4.4.6
WE 17.14 Intermediate transverse stiffeners
For panels BC and CD a = 3900 mm and tmin = 8 mm (see WE 17.8).
Try stiffener consisting of two flats 80 ¥ 15 mm
ps = 275 N/mm2
e=1
bs = 80 < 13 e ts = 13 ¥ 1 ¥ 13 = 169 mm
OK
4.5.1.2
4.4.6.4
Check minimum stiffness
a 3900
=
= 1.95 > 2
d 2000
3
Is = 0.75d t min
= 0.75 ¥ 2000 ¥ 83
= 0.768 ¥ 106 mm4
Actual Iy of stiffener neglecting web
Iy =
15 ¥ 175 3
= 6.70 ¥ 10 6 mm 4
12
Iy = 6.70 ¥ 106 > Is = 0.768 ¥ 106 mm4
Check buckling resistance of stiffeners
OK
4.4.6.6
Steel Designers' Manual - 6th Edition (2003)
Worked examples
Subject
The
Steel Construction
Institute
Chapter ref.
PLATE GIRDERS
17
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
Made by TMR
Checked by
Sheet no.
11
BD
Intermediate transverse stiffeners should be designed for a compressive axial force Fq given by
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
Fq = Vmax - Vcr
For panels BC and CD (see WE 17.12)
Vmax = 2206 kN
Vcr = 1983 kN
Fq = 2206 - 1983 = 223 kN
Since Fq is less than the external load applied to the stiffener at D
see WE 17.15 for design.
WE 17.15 Intermediate load-carrying stiffener at D
This stiffener is subjected to an external compressive force Fx =
920 kN applied in line with the web.
Try stiffener consisting of two flats 80 ¥ 15 mm.
Minimum stiffness requirement as for other transverse stiffeners is
satisfied (see WE 17.14).
Check buckling resistance assuming a width of web equal to 15t
on each side of the centreline of the stiffener (see Fig. WE 17.4).
225
225
80
jri:
15
Fig.WE 17.4
Load-carrying stiffener at D
4.5.3.3
507
Steel Designers' Manual - 6th Edition (2003)
Worked examples
508
Subject
The
Steel Construction
Institute
Chapter ref.
PLATE GIRDERS
17
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
Made by TMR
Checked by
Sheet no.
12
BD
For panels CD and DE
(Vcr)CD = 1983 kN
(Vcr)DE = 1778 kN
H.2
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
For both panels Fq = Vmax - Vcr is negative and therefore < Fx =
920 kN. Stiffener should therefore be designed for
Fx < Px = Aspc
4.4.6.6
4.5.3.3
4.7.5
As = 160 ¥ 15 + 30 ¥ 15 ¥ 15 = 9150 mm
2
Neglecting the small contribution of the web:
15 ¥ 175 3
= 6.70 ¥ 10 6 mm 4
12
Iy =
Iy ˆ
ry = Ê
Ë As ¯
12
6
Ê 6.70 ¥ 10 ˆ
=Á
˜
Ë
9150 ¯
12
= 27.06 mm
LE = 0.7 ¥ 2000 = 1400 mm
l=
LE
1400
=
= 51.73
ry
27.06
pys = 275 N /mm2
pc = 249 N/mm2
Table 24
Px = Aspc = 9150 ¥ 249 ¥ 10 - 3 = 2278 kN
Fx = 920 < Px = 2278 kN
OK
WE 17.16 Loads applied between stiffeners
The web of the plate girder is subjected to a factored UDL of
92 kN/m applied between stiffeners, which corresponds to a stress
fed given by
fed =
92
= 6.13 N /mm2
15
When the compression flange is restrained against rotation relative
to the web
2 ˘ E
È
ped = Í 2.75 +
(a d )2 ˙˚ (d t )2
Î
4.5.3.2
Steel Designers' Manual - 6th Edition (2003)
Worked examples
Subject
The
Steel Construction
Institute
Chapter ref.
PLATE GIRDERS
17
Silwood Park, Ascot, Berks SL5 7QN
Design code
BS 5950: Part 1
Made by TMR
Checked by
Sheet no.
13
BD
Check for largest panel only
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
2
È
˘ 205000
ped = Í 2.75 +
= 34.6 N /mm2
2˙
(5650 2000 ) ˚ ( 2000 15 )2
Î
fed = 6.13 < ped = 34.6 kN/mm2
OK
WE 17.17 Stiffener bounding anchor panel at B
Since the compressive force Fs due to the support reaction is zero,
the stiffener should be checked for a tensile force equal to Ftf
Ftf = 0.15Hq
Hq = 3193 kN (see WE 17.11)
d = 2000 mm
ae = 1550 mm
Ftf = 0.15 ¥ 3193 ¥
2000
= 618 kN
1550
Since this force is significantly less than the compressive resistance
of the transverse stiffeners (see WE 17.15) stiffener is adequate.
H.4.4
509
Steel Designers' Manual - 6th Edition (2003)
510
Worked examples
Subject
The
Steel Construction
Institute
Chapter ref.
PLATE GIRDERS
17
Silwood Park, Ascot, Berks SL5 7QN
Made by TMR
Design code
BS 5950: Part 1
Checked by
BD
WE 17.18 Final girder
AL! intermedicite sti ffeners
2x80x15
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
2x280x22
Nfl
1550
I
I
3900 3900
'-1 h
I
I
5550
I
30000
750
2100
_____50
Fig.WE 17.5
Final plate girder details
H
Sheet no.
14
Steel Designers' Manual - 6th Edition (2003)
Chapter 18
Members with compression
and moments
by DAVID NETHERCOT
This material is copyright - all rights reserved. Reproduced under licence from The Steel Construction Institute on 12/2/2007
To buy a hardcopy version of this document call 01344 872775 or go to http://shop.steelbiz.org/
18.1 Occurrence of combined loading
Chapters 15 and